Lincoln University Digital Dissertation Copyright Statement The digital copy of this dissertation is protected by the Copyright Act 1994 (New Zealand). This dissertation may be consulted by you, provided you comply with the provisions of the Act and the following conditions of use: 


you will use the copy only for the purposes of research or private study you will recognise the author's right to be identified as the author of the dissertation and due acknowledgement will be made to the author where appropriate you will obtain the author's permission before publishing any material from the dissertation. Animal liveweight gain from lucerne and
lucerne/grass mixes.
A dissertation/thesis
submitted in partial fulfillment
of the requirement for the Degree of
Bachelor of Agricultural Science with Honours
at Lincoln University
by
E. J. Coutts
Lincoln University
2013
i
ABSTRACT
Abstract of a Thesis submitted in partial fulfillment of the requirement for
the Degree of Bachelor of Agricultural Science with Honours
Animal liveweight gain from lucerne and lucerne-grass mixes.
By
E. J. Coutts
Animal production from lucerne has proved superior to other pastures in dryland
situations, however there is interest in lucerne/grass mixes an as alternative to lucerne
monocultures. A farmlet study was set up with lucerne monocultures, lucerne/cocksfoot
and lucerne/brome mixes. Annual liveweight production was 865 kg/ha for lucerne
monocultures and lucerne/cocksfoot, compared with 746 kg/ha for lucerne/brome.
During spring, 78% of total liveweight production was from lambs that maintained growth
rates of ∼300 g/head/d on all pastures. Accumulated dry matter yields were ∼12 t/ha
for all pastures. All pastures used 612 mm of water at an efficiency of 22 kg DM/ha/mm.
Temperature did not influence dry matter yield across pastures, all produced 5.5 kg
DM/ha/°Cd. Botanical composition was similar for all pastures, but livestock on lucerne
monocultures consumed 100 kg DM/ha more weeds than lucerne/grass. Stock grazing
lucerne/brome actively selected for legumes, and against brome and dead material,
resulting in a decline in pasture quality. Differences in liveweight gain were explained by
different stocking rates for each treatment, not by quantity or quality of herbage.
Utilizing a ‘leader follower’ system of ewes and lambs followed by cattle could aid to
maintain the herbage quality of lucerne/grass mixes by avoiding a built up of
reproductive and dead material.
Keywords: botanical composition, brome, Bromus willdenowii, cocksfoot, Dactylis
glomerata, dryland, grazing brome, Medicago sativa, prairie grass, water use,
i
TABLE OF CONTENTS
ABSTRACT ............................................................................................................................... i
Table of Contents .................................................................................................................. ii
List of Tables .......................................................................................................................... v
List of Figures ........................................................................................................................ ix
List of Plates .......................................................................................................................... xi
List of Appendices................................................................................................................ xii
1
INTRODUCTION ............................................................................................................. 1
2
REVIEW OF THE LITERATURE ......................................................................................... 3
2.1
Pasture species ....................................................................................................... 3
2.2
Animal production .................................................................................................. 4
2.3
Animal requirements .............................................................................................. 8
2.4
Quality ..................................................................................................................... 9
2.5
Grazing selection................................................................................................... 12
2.6
Annual dry matter production .............................................................................. 13
2.7
Botanical composition .......................................................................................... 15
2.8
Seasonal production ............................................................................................. 17
Water use ........................................................................................................................ 19
2.9
3
Conclusions ........................................................................................................... 20
MATERIALS AND METHODS......................................................................................... 21
3.1
Experimental site .................................................................................................. 21
3.2
Experimental area ................................................................................................. 21
3.3
Experimental design ............................................................................................. 22
3.4
Soil fertility ............................................................................................................ 23
3.5
Fertiliser ................................................................................................................ 24
3.6
Meteorological data.............................................................................................. 24
3.7
Soil water budget .................................................................................................. 26
3.7.1
Potential soil water deficit............................................................................. 26
3.7.2
Soil water content (SWC)............................................................................... 27
3.7.3
Water use efficiency ...................................................................................... 27
3.8
Livestock and grazing management ..................................................................... 28
3.9
Weed control ........................................................................................................ 30
3.10
Measurements .................................................................................................. 30
ii
3.10.1 Live weight measurements............................................................................ 31
3.10.2 Dry matter measurements ............................................................................ 31
3.10.3 Lucerne quality .............................................................................................. 34
3.10.4 Thermal time ................................................................................................. 35
3.10.5 Selection of the base temperature................................................................ 35
3.10.6 Statistical analysis .......................................................................................... 36
4
RESULTS ....................................................................................................................... 38
4.1
Animal production ................................................................................................ 38
4.1.1
Annual liveweight production ....................................................................... 38
4.1.2
Rotational liveweight production from ewes and lambs .............................. 39
4.1.3
Total graze days ............................................................................................. 40
4.1.4
Investigating the effect of pasture age ......................................................... 41
4.1.5
Ewe liveweight ............................................................................................... 41
4.1.6
Rotational animal liveweight gain ................................................................. 42
4.2
Pasture dry matter yield ....................................................................................... 44
4.2.1
Accumulated dry matter yield ....................................................................... 44
4.2.2
Mean daily growth rates ............................................................................... 45
4.3
Thermal time relationships ................................................................................... 46
4.4
Botanical composition .......................................................................................... 48
4.4.1
Annual botanical composition ....................................................................... 48
4.4.2
Pasture botanical composition in different grazing rotations ...................... 49
4.5
4.4.2.1
Rotation 1 ............................................................................................... 49
4.4.2.2
Rotation 2 ............................................................................................... 51
4.4.2.3
Rotation 3 ............................................................................................... 53
4.4.2.4
Rotation 5 ............................................................................................... 55
Nutritive yield ....................................................................................................... 57
4.5.1
Metabolisable energy yield ........................................................................... 57
4.5.1.1
Annual metabolisable energy yield ........................................................ 57
4.5.1.2
Grazing Rotation 1 metabolisable energy yield ..................................... 57
4.5.1.3
Grazing Rotation 2 metabolisable energy yield ..................................... 58
4.5.1.4
Grazing Rotation 3 metabolisable energy yield ..................................... 59
4.5.1.5
Grazing Rotation 5 metabolisable energy yield ..................................... 60
4.5.2
Nitrogen (N) yield .......................................................................................... 61
4.5.2.1
Annual nitrogen yield ............................................................................. 61
4.5.2.2
Grazing Rotation 1 nitrogen yield .......................................................... 61
iii
4.6
5
Grazing Rotation 2 nitrogen yield .......................................................... 62
4.5.2.4
Grazing Rotation 3 nitrogen yield .......................................................... 62
4.5.2.5
Grazing Rotation 5 nitrogen yield .......................................................... 64
Soil water content ................................................................................................. 65
4.6.1
Available water .............................................................................................. 65
4.6.2
Soil water content ......................................................................................... 67
4.6.3
Water use efficiency ...................................................................................... 68
DISCUSSION ................................................................................................................. 70
5.1
Animal production ................................................................................................ 70
5.1.1
Annual and seasonal liveweight production ................................................. 70
5.1.2
Rotational liveweight gains ........................................................................... 71
5.2
Pasture dry matter yield ....................................................................................... 72
5.2.1
Accumulated dry matter yield ....................................................................... 72
5.2.2
Mean daily growth rates ............................................................................... 73
5.3
Soil water budget .................................................................................................. 73
5.4
Thermal time ......................................................................................................... 74
5.5
Botanical composition .......................................................................................... 74
5.6
Nutritive yield ....................................................................................................... 75
5.6.1
Metabolisable energy .................................................................................... 75
5.6.2
Nitrogen yield ................................................................................................ 77
5.6.3
Production graze days ................................................................................... 77
5.7
6
4.5.2.3
Conclusions ........................................................................................................... 79
GENERAL DISCUSSION anD CONCLUSIONS ................................................................. 80
6.1
General discussion ................................................................................................ 80
6.1.1
Establishment ................................................................................................ 80
6.1.2
Sodium content ............................................................................................. 80
6.1.3
Effect of nitrogen on pastures ....................................................................... 81
6.1.4
Grazing management .................................................................................... 82
6.2
Conclusions ........................................................................................................... 84
7
Acknowledgements ..................................................................................................... 85
8
References ................................................................................................................... 86
9
Appendices .................................................................................................................. 89
iv
LIST OF TABLES
Table 2.1 Crude protein (g/g DM) and ME (MJ/kg DM) contents of herbage fractions
(palatable, unpalatable and weed) and post-grazing residual of chicory,
lucerne and red clover swards grown over five years. Values in parenthesis
are standard errors for each mean calculated from five year’s data. Adapted
from Brown et al. (2004). ................................................................................. 11
Table 2.2 Annual herbage yields of lucerne and lucerne/prairie grass with three sowing
rates over three years. Adapted from Baars and Cranston (1978). ................. 14
Table 2.3 Seasonal production of lucerne and lucerne prairie grass pastures. Adapted
from McKenzie et al. 1990................................................................................ 18
Table 3.1 Species cultivar, sowing date, rate and drill type for paddocks C6E, C7W and C7E
at Ashley Dene, Canterbury. ............................................................................. 22
Table 3.2 Soil test results from May/June 2011 for paddocks C6E, C7W and C7E, Ashley
Dene, Canterbury, New Zealand. ..................................................................... 23
Table 3.3 Summary of stock class, start and end date and plots grazed for each grazing
rotation at Ashley Dene, Canterbury, New Zealand. Where E & L denotes ewes
and lambs, W L denotes weaned lambs, Ram Hgts denotes ram hoggets and
Ewe Hgts denotes ewe hoggets........................................................................ 29
Table 3.4 Grazing rotation stocking rates (SR), expressed as stock units (SU/ha) for lucerne
monocultures, lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br)
pastures for either production or maintenance liveweight (LWT) from 5/09/12
to 30/06/13 at Ashley Dene, Canterbury, New Zealand. ................................. 29
Table 3.5 Summary of stock classes and measurements taken for grazing rotations from
1/07/12 to 30/06/13. Where E & L denotes ewes and lambs, W L denotes
weaned lambs, Ram Hgts denotes ram hoggets and Ewe Hgts denotes ewe
hoggets. Measurements taken include dry matter yield (DM), animal
liveweight gain (LWt), botanical composition (BC) and nutritive value (NU) and
are indicated by a ‘Y’. LWT period determines if the rotation was a
‘production’ or ‘maintenance’ period. Liveweight (LWT) rotation is when stock
were weighed relative to grazing rotations. .................................................... 30
Table 4.1 Total, maintenance and production graze days (GD/ha) for lucerne
monocultures, lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes
from 1/07/12 to 30/06/13 at Ashley Dene, Canterbury, New Zealand.
Production GD are for rotations when liveweight was measured and
maintenance GD are for rotations when liveweight was not measured. ........ 41
Table 4.2 Liveweight gain (g/head/d) of ewes and lambs grazed on lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes over liveweight
Rotations 1, 2 and 3 from 5/09 to 23/11/12 at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 43
Table 4.3 Liveweight gain (g/head/d) of weaned lambs grazing lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) pastures over
v
liveweight Rotation 4 from 28/11/12 to 4/01/13 at Ashley Dene, Canterbury,
New Zealand. .................................................................................................... 43
Table 4.4 Liveweight (LWT) gain (g/head/d) of ewe hoggets grazing lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) pastures over
liveweight Rotation 5 from 15/05 to 18/06/13. ............................................... 44
Table 4.5 Annual pre-grazing botanical composition of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 48
Table 4.6 Annual post-grazing botanical composition of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 49
Table 4.7 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for grazing rotation one from 5/09
to 16/10/12 prior to grazing with ewes and lambs at Ashley Dene, Canterbury,
New Zealand. .................................................................................................... 50
Table 4.8 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for grazing rotation one from
5/09/ to 16/10/12 after grazing with ewes and lambs at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 50
Table 4.9 Dry matter (kg DM/ha) consumed by ewes and lambs of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes during grazing
rotation one from 5/09 to 16/10/2012 at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 51
Table 4.10 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 2 from 24/10 to
23/11/12 prior to grazing with ewes and lambs at Ashley Dene, Canterbury,
New Zealand. .................................................................................................... 52
Table 4.11 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for Rotation 2 from 24/10 to
23/11/12 after grazing with ewes and lambs at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 52
Table 4.12 Dry matter (kg DM/ha) consumed by ewes and lambs of lucerne
monocultures, lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) pastures
during Rotation 2 from 24/10 to 23/11/2012 at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 53
Table 4.13 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for grazing Rotation 3 from
28/11/12 to 4/01/13 prior to grazing with weaned lambs at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 53
Table 4.14 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for grazing Rotation 3 from
28/11/12 to 4/01/13 after grazing with weaned lambs at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 54
Table 4.15 Dry matter (kg DM/ha) consumed by weaned lambs of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes during grazing
vi
Rotation 3 from 28/11/2012 to 4/01/2013 at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 54
Table 4.16 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 5 from 15/5 to
18/06/13 prior to grazing with ewe hoggets at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 55
Table 4.17 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 5 from 15/5 to
18/06/13 after grazing with ewe hoggets at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 56
Table 4.18 Dry matter (kg DM/ha) consumed by ewe hoggets of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes during Rotation 5
from 15/5 to 18/06/2013 at Ashley Dene, Canterbury, New Zealand. ............ 56
Table 4.19 Annual metabolisable energy (MJ ME/ha) and corresponding yield (GJ ME/ha)
and corresponding ME values (MJ/ha) of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes at Ashley Dene,
Canterbury. Sown species yield is presented as the sum of the lucerne and
sown grass yields. ............................................................................................. 57
Table 4.20 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) and corresponding ME values (MJ/ha) of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 1
from 5/09 to 24/10/12 prior to grazing with ewes and lambs at Ashley Dene,
Canterbury, New Zealand. Sown species yield is presented as the sum of
lucerne and sown grass yields. ......................................................................... 58
Table 4.21 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) and corresponding ME values (MJ/ha) of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes for grazing
Rotation 2 from 24/10 to 28/11/12 prior to grazing with ewes and lambs at
Ashley Dene, Canterbury, New Zealand. Sown species yield is presented as the
sum of lucerne and sown grass yields. ............................................................. 58
Table 4.22 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) and of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for Rotation 3 from 28/11/12 to 4/01/13 prior
to grazing with weaned lambs at Ashley Dene, Canterbury, New Zealand. .... 59
Table 4.23 Post-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) of lucerne, lucerne/cocksfoot and lucerne/brome pastures for grazing
Rotation 3 from at Ashley Dene, Canterbury, New Zealand. ........................... 59
Table 4.24 Metabolisable energy consumed by weaned lambs grazing lucerne,
lucerne/cocksfoot and lucerne/brome pastures in early summer at Ashley
Dene, Canterbury, New Zealand. ..................................................................... 60
Table 4.25 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 5 at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 60
vii
Table 4.26 Annual nitrogen concentration (N%) and corresponding annual nitrogen yield
(GJ ME/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) pastures at Ashley Dene, Canterbury. Sown species
yield is presented as the sum of lucerne and sown grass yields. ..................... 61
Table 4.27 Pre-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 1 at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 61
Table 4.28 Pre-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 2 at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 62
Table 4.29 Pre-grazing nitrogen concentration (N%) and nitrogen yield (kg/ha) of lucerne
monocultures, lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes
for Rotation 3 from 28/11/12 to 4/01/13 prior to grazing with weaned lambs
at Ashley Dene, Canterbury, New Zealand. ...................................................... 63
Table 4.30 Post-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 3 from 28/11/12 to
4/01/13 after grazing with weaned lambs at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 63
Table 4.31 Nitrogen consumed (kg N/ha) by weaned lambs grazing lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes during Rotation
3 from 28/11/12 to 4/01/13 at Ashley Dene, Canterbury, New Zealand. ....... 64
Table 4.32 Pre-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 5 from 15/5 to 18/06/13
prior to grazing with ewe hoggets at Ashley Dene, Canterbury, New Zealand.
.......................................................................................................................... 64
Table 4.33 Water use (WU) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes from 1/07/12 to 30/06/13 at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 67
viii
LIST OF FIGURES
Figure 2.1 Liveweight produced per hectare from six dryland pastures at Lincoln
University. Cf = cocksfoot, Bc = balansa clover, Sc = subterranean clover, Cc =
Caucasian clover, Wc = white clover, Rg = ryegrass, Luc = lucerne. Error bars
represent one LSD above periods when production was different. From Brown
et al. (2006). ........................................................................................................ 5
Figure 2.2 Annual dry matter production in a) 2004/05 and b) 2005/06 of six dryland
pastures grown at Lincoln University. Cf = cocksfoot, Cc = balansa clover, Sc =
subterranean clover, Cc = Caucasian clover, Wc = white clover, Rg = ryegrass,
Luc = lucerne. Bars represent one LSD. From Brown et al. (2006)..................... 7
Figure 2.3 Botanical composition of lucerne and lucerne/prairie grass pastures under low,
medium and high stocking rates. Where LSR denotes 5.0 cattle/ha, MSR
denotes 6.67 cattle/ha and HSR denotes 10.0 cattle/ha. From Marsh and
Brunswick (1977). ............................................................................................. 16
Figure 3.1 Mean monthly rainfall (a) and air temperature (b) for the 2012/2013 growing
season with long term means for the period 1975-2010 (air temperature) and
1980-2009 (rainfall). Air temperature data were obtained from Broadfields
meteorological station (43°62’S, 172°47’E). Rainfall data were obtained from
Ashley Dene weather station (43°65’S,172°35’E). ........................................... 25
Figure 3.2 Potential soil moisture deficit (PSMD, mm) between 01/07/2012 and
31/05/2013 for paddocks C6E, C7W and C7E at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 26
Figure 3.3 Pre (a) and post-graze (b) linear regressions of lucerne height versus dry matter
(DM) yield for spring (●), summer (○), and autumn (▽) at Ashley Dene,
Canterbury. Forms of the regression were: Spring/summer pre-graze yield =
261±136 + 84±4.2x R2=0.70, autumn pre-graze yield = 510±130 + 42±6.5x
R2=0.43. Post-graze spring/summer/autumn yield = 198±85.6 + 91.4±4.36x
R2=0.64.............................................................................................................. 34
Figure 3.4 Base temperatures (Tb) and corresponding R2 values for lucerne monocultures
(●), lucerne/cocksfoot (■) and lucerne/brome (▽) pastures using a twostage model for spring 2012 data. The gray area indicates the R2 values for a
three-stage model with a Tb=1°C...................................................................... 36
Figure 4.1 Annual liveweight production of lucerne monocultures, lucerne/brome
(Luc/Br) and lucerne/cocksfoot (Luc/CF) mixes over five liveweight production
periods from 1/07/2012 to 30/06/2013 at Ashley Dene, Canterbury, New
Zealand. Stacked bars represent spring liveweight gain with ewes and lambs
(■), summer liveweight gain with weaned lambs (▩) and autumn liveweight
gain with ewe hoggets (■). The error bar is SEM for accumulated liveweight
production. ....................................................................................................... 39
Figure 4.2 Spring liveweight production (kg LWT/ha) for rotation one for ewes (a) and
lambs (c) and rotation two ewes (b) and lambs (d) grazing lucerne
monocultures (■), lucerne/cocksfoot (▩) and lucerne/brome (■) mixes at
ix
Ashley Dene, Canterbury, New Zealand. The error bars are SEM for liveweight
production across treatments. ......................................................................... 40
Figure 4.3 Change in lactating ewe liveweight over two dry matter rotations from 5/09 to
24/11/12 on lucerne monocultures (●), lucerne/cocksfoot (■) and
lucerne/brome (▽) mixes at Ashley Dene, Canterbury, New Zealand............ 42
Figure 4.4 The total accumulated dry matter (DM) yield of lucerne monocultures (●),
lucerne/brome (▽), and lucerne/cocksfoot (■) pastures from 1/07/2012 to
30/06/2013 at Ashley Dene, Canterbury, New Zealand. Grey area indicates
the period when no measurements were taken due to low summer growth. 45
Figure 4.5 Mean daily growth rates of lucerne monocultures (●), lucerne/cocksfoot (■)
and lucerne/brome (▽) mixes for regrowth cycles between 1/07/12 and
30/06/13 at Ashley Dene, Canterbury, New Zealand. Error bars are SEM for
each harvest date. Vertical grey bars indicate maintenance grazing periods
where no dry matter measurements were taken. ........................................... 46
Figure 4.6 Relationship between accumulated dry matter (DM) yield and accumulated
thermal time (°Cd, Tb=0°C) for lucerne monocultures (●), lucerne/cocksfoot
(■) and lucerne/brome (▽) mixes. Forms of the spring regression lines were:
Yield = 5.5±0.19x – 1933±281 (R2=0.99). Thermal time was accumulated using
air temperature. Grey lines extrapolate back to the x-intercept. Full details of
regression in Appendix 4. ................................................................................. 47
Figure 4.7 Water extraction pattern of lucerne, cocksfoot and brome roots in the soil
profile. Where (●) is the upper limit and (○) is the lower limit for plant
available water in Plot 7 (top) and Plot 2 (bottom) in paddocks C6E and C7W,
Ashley Dene, Canterbury. ................................................................................. 66
Figure 4.8 Soil water content (mm) and rainfall (mm) for Plot 7 (—) and Plot 2 (—) in
paddocks C6E and C7W, Ashley Dene, Canterbury. Rainfall data are taken
from the Ashley Dene weather station (43°65’S, 172°32’E). ........................... 68
Figure 4.9 Relationship between accumulated dry matter yield (kg DM/ha) and
accumulated water use (mm) for lucerne monocultures (●),
lucerne/cocksfoot (■) and lucerne/brome (▽) mixes. Form of the spring
regression line is: 22.0±0.11x + 85.5±29.6 (R2=0.99). ...................................... 69
x
LIST OF PLATES
Plate 1 Map of experimental design showing paddocks C6E, C7W and C7E and plots 1-18
at Ashley Dene, Canterbury. The total experimental area is 17.7 ha. ............. 23
Plate 2 Ewe hoggets grazing lucerne/brome pastures on 27 May 2013 at Ashley Dene,
Canterbury, New Zealand. Reproductive stems are visible, highlighting the
quality decline................................................................................................... 32
Plate 3 Botanical composition of a lucerne/cocksfoot pasture on 16 May 2012 from
Ashley Dene, Canterbury, New Zealand. Botanical components are lucerne,
cocksfoot, weeds and dead material................................................................ 33
xi
LIST OF APPENDICES
Appendix 1 Daily metabolisable energy requirements (MJ ME) of ewes for maintainance
and various stages of preganancy and lactation. From Nicol and Brookes
(2007). ............................................................................................................... 89
Appendix 2 Soil map of paddocks C6E, C7W and C7E at Ashley Dene, Canterbury, New
Zealand. ............................................................................................................ 90
Appendix 3 Detailed stock movements for grazing rotations on lucerne monocultures,
lucerne/cocksfoot and lucerne/brome mixes at Ashley Dene, Canterbury, New
Zealand from 1/07/12 to 30/06/13. ................................................................. 91
Appendix 4 Regression equations and coefficients of determination for the regression of
cumulated thermal against cumulated dry matter yield time in spring 2012 of
lucerne monocultures, lucerne/cocksfoot and lucerne/brome mixes at Ashley
Dene, Canterbury, New Zealand. ..................................................................... 94
Appendix 5 Regression equations and coefficients of determination for the regression of
cumulated water use (WU) against cumulated dry matter yield in spring 2012
of lucerne monocultures, lucerne/cocksfoot and lucerne/brome mixes at
Ashley Dene, Canterbury, New Zealand. .......................................................... 95
Appendix 6 Water extraction pattern of lucerne and brome roots in the soil profile.
Where (●) is the upper limit and (○) is the lower limit for plant available
water in plots 1,3-6,8-18 in paddocks C6E, C7W and C7E at Ashley Dene,
Canterbury, New Zealand. ................................................................................ 96
xii
1
INTRODUCTION
Perennial ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.) are the most
commonly sown pasture in New Zealand. These pastures are productive with adequate
rainfall or irrigation. However, both species have shallow roots, reducing their ability to
access water (Brown et al., 2006). Evapotranspiration often exceeds rainfall during
summer months in dryland regions, resulting in unreliable production from these
traditional pastures (Moot, 2012). This is a disadvantage for dryland sheep systems, with
high feed demand in late spring and summer months for finishing lambs (Fraser et al.,
1999). Spring is the most important time to maximise pasture production, as moist soil
and low evapotranspiration rates reduce water stress, allowing reliable pasture
production (Brown et al., 2006). Alternative pasture species such as lucerne (Medicago
sativa L.) have been utilised in dryland areas due to their ability to tolerate drought
conditions.
Lucerne has been promoted in dryland farming systems in New Zealand for over 100
years, however the area of lucerne has decreased since 1975 (Douglas, 1986; Moot,
2012). A survey completed in 2000 showed that 67% of dryland farmers in the South
Island grew lucerne but it only averaged 17% of their farm area (Kirsopp, 2001). This is
despite a recommendation from White (1982) that 40-60% of the farm should be in
lucerne to maximize liveweight gain. Traditionally, farmers have been advised to wait
until 10% flowering in lucerne stands before grazing. Now, it is recommended that the
first paddock of a lucerne rotation is grazed at 1500 kg DM/ha with ewes and two week
old lambs (Moot, 2012). The change in grazing management has increased lucerne use in
dryland systems. Lucerne has been successful, particularly for dryland properties in
Marlborough (Avery et al., 2008). Incorporation of lucerne into farming systems has
adapted properties to drier conditions and increased profitability, with lucerne providing
a reliable feed source during drought.
The growth pattern of lucerne is a limitation due to it being dormant during winter
months, therefore production is low. Lucerne has a reputation for slow early spring
1
growth (Tonmukayakul et al., 2009) therefore farming systems with lucerne tend to
require later lambing dates. Later lambing is perceived to restrict lamb growth resulting
in less animals being finished before the onset of summer drought (Moot, 2012).
This limitation can be overcome by sowing a more winter active species such as cocksfoot
(Dactylis glomerata L.) or prairie grass (Bromus willdenowi Kunth.) which complement the
lucerne growth curve (O'Connor, 1967; Fraser and Vartha, 1979; Fraser, 1982). There was
a lot of research carried out on the sociability of lucerne during the 1970’s and 1980’s
however; there is little recent literature on the performance of lucerne-grass mixes.
Advantages of lucerne-grass mixes are that grass species provide herbage during winter
when the lucerne is dormant, which reduces the interspecific competition for resources
(McKenzie et al., 1990). Sowing a grass species may also decrease the weed content of
the lucerne stand (Cullen, 1965). Intensive grazing management for lucerne-grass mixes is
vital to ensure that the stand does not become dominated by one species.
The primary objective of this study is to compare animal liveweight gain from lucerne and
two lucerne-grass mixes – lucerne/cocksfoot and lucerne/brome. Annual pasture
production and quality over the 2012/13 growing season will be used to explain
differences in liveweight gain. Differences in quantity will be explained by annual yield,
completing a soil water budget and determining thermal time requirements. Differences
in quality will be explained by botanical composition pre and post graze and nutritive
analysis of samples taken throughout the experimental period.
This dissertation is presented in six chapters with a review of the literature, materials and
methods, results followed by a discussion and conclusions. A general discussion relates
the research back to wider farming systems.
2
2
REVIEW OF THE LITERATURE
2.1 Pasture species
Lucerne is a perennial legume species commonly grown in drought prone areas
throughout New Zealand (Wigley et al., 2012). It is known for high dry matter production
and high quality feed in dryland conditions. Annual dry matter production of Kaituna
lucerne was 28 t/ha under dryland conditions on a Wakanui silt loam (Moot et al., 2008).
Lucerne can either be grazed by livestock or cut and carried for supplementary feed.
A major advantage of lucerne is its long taproot allowing it to extract water from deeper
in the soil profile and use it more efficiently than grass species. The taproot can penetrate
1.5-2.0 m in the first season and 9-12 m at maturity depending on the soil structure. In an
experiment by Moot et al. (2008) lucerne extracted 328 mm of water to a depth of 2.3 m
while perennial ryegrass extracted only 243 mm to a depth of 1.5 m on the same soil.
Extracting more water than perennial ryegrass results in lucerne having greater access to
water but it also uses each mm more efficiently.
Lucerne has the ability to fix nitrogen due to a symbiotic relationship with rhizobia
bacteria Ensifer meliloti (Wynn-Williams, 1982). The fixation of atmospheric nitrogen
makes it plant available, allowing companion grass species to utilise it for production.
A potential disadvantage of lucerne is its need for rotational grazing due to the growing
point being at the top of the plant and the crown being the main region of regrowth
(Langer and Keoghan, 1970). Lucerne is also dormant during winter months, therefore
dry matter production is low. Low winter production can be overcome by sowing a winter
active grass species such as prairie grass or drought tolerant cocksfoot to complement
the lucerne growth curve (O'Connor, 1967; Fraser and Vartha, 1979; Fraser, 1982). The
difficulty with lucerne/grass mixes is finding a combination that doesn’t become one
species dominant due to interspecific competition for resources.
Cocksfoot is a tufted perennial grass species that is well adapted to moderate fertility and
drought conditions (Barker et al., 1985). It is the most commonly sown grass species after
3
perennial ryegrass (Vartha, 1975). It is a slow establishing species, however once
established cocksfoot has aggressive growth, often dominating the pasture sward.
Cocksfoot is frequently nitrogen deficient and therefore has low palatability (Mills et al.,
2006).
Prairie grass is a perennial grass with broad leaves and a sparse tiller production. It has an
upright growth habit which is intolerant of continuous grazing. It is palatable throughout
the year and its main value is as a cool-season grass (O'Connor, 1967).
This literature review will describe the animal performance reported on lucerne, animal
requirements for production at different stages of their life and how the quality of
lucerne pastures may meet these animal demands. The annual and seasonal dry matter
yield and botanical composition of lucerne/grass mixes and the water use will also be
presented. The review will focus on cocksfoot and prairie grass as companion species for
lucerne crops as these were the species used in this experiment.
2.2 Animal production
The ultimate test of pasture quality is its effect on livestock production. Brown et al.
(2006) investigated the temporal pasture and livestock production of six dryland pastures
in ‘Maxclover’ over two years at Lincoln University. The pasture treatments were
cocksfoot with balansa clover (Trifolium michelianum (Cf/Bc), cocksfoot with Caucasian
clover (T. ambiguum (Cf/Cc), cocksfoot with subterranean clover (T. subterraneum
(Cf/Sc), cocksfoot with white clover (T. repens L. (Cf/Wc), ryegrass (Lolium perenne L.)
with white clover (Rg/Wc) and lucerne (Luc). Figure 2.1 shows the liveweight gain
produced (kg/ha) over spring, summer and autumn for 2004/05 and 2005/06. In spring
2004, production for all grass treatments averaged 300 kg LW/ha except Cf/Cc which
produced 200 kg LW/ha. Lucerne had the highest (P<0.001) production for spring 2004
with 400 kg LW/ha, even though grazing of the lucerne treatments commenced 40 days
later than the grass treatments. Lucerne also had the highest (P<0.001) summer
production in 2004 with approximately 550 kg LW/ha. Autumn production was low for all
mixes with Rg/Wc producing the greatest (P<0.001). In spring 2005 Cf/Sc had the highest
(P<0.001) production while lucerne and Cf/Cc produced the least. The low liveweight
4
production from lucerne was attributed to a dry winter in 2005 with only 50 mm of
rainfall for June, July and August which was much lower than the long term average of 60
mm of rainfall per month. This resulted in the advantages of the lucerne taproot not
being expressed.
Figure 2.1 Liveweight produced per hectare from six dryland pastures at Lincoln
University. Cf = cocksfoot, Bc = balansa clover, Sc = subterranean clover, Cc =
Caucasian clover, Wc = white clover, Rg = ryegrass, Luc = lucerne. Error bars
represent one LSD above periods when production was different. From Brown
et al. (2006).
Liveweight production can be explained by the combination of quality and quantity of
pasture grown. Lucerne produced the most (P<0.001) liveweight in 2004/05 and also the
most (P<0.001) dry matter (Figure 2.2). In 2004/05 Cf/Sc and Cf/Wc were the most
(P<0.001) productive grass based pasture mixes and Cf/Sc was the most (P<0.001)
productive in 2005/06. Growth rates of the grass-clover mixes ranged from < 10 kg/ha/d
during March-July up to a maximum of 60-105 kg/ha/d in October. Cf/Sc consistently had
the highest (P<0.05) growth rates from August-October. The rate of clover growth
increased from June to a maximum in October. The Cf/Sc pasture had a growth rate of
2.3 kg/ha/d in June, which rose to a maximum of 63 kg/ha/d in October. The high yields
5
and legume content of the Cf/Sc resulted in greater quality and quantity of the pasture,
which contributed to the higher liveweight production.
The grazing management of the experiment is important to consider when determining
the legume content of the pastures. Frequent grazing intervals will favour grass growth
while infrequent grazing intervals favour legume growth. In the ‘Maxclover’ experiment,
grazing management was aimed to maximise stock production by using the optimal
grazing management for each pasture type (Brown et al., 2006). All treatments were
rotationally grazed at three grazing intervals – short, medium or long. A ‘put and take’
system was used with sheep grazing the trial. There was a ‘core’ group of animals in the
liveweight gain trial plus ‘spares’ which were used to match feed demand with supply
throughout each grazing period. Lucerne grazing began later than grass treatments in
both years, due to the later growth pattern of lucerne compared with grass species. Hard
grazing was often used in cocksfoot treatments to maintain palatable vegetative growth
and an open pasture allowing legume species to remain productive. Brown et al. (2006)
concluded that no single pasture proved consistently superior but a combination of
species could be used to complement each other as each species showed strengths at
different times of the year under different climatic conditions.
6
Figure 2.2 Annual dry matter production in a) 2004/05 and b) 2005/06 of six dryland
pastures grown at Lincoln University. Cf = cocksfoot, Cc = balansa clover, Sc =
subterranean clover, Cc = Caucasian clover, Wc = white clover, Rg = ryegrass,
Luc = lucerne. Bars represent one LSD. From Brown et al. (2006).
Mills et al. (2008b) reported on the same experiment as Brown et al. (2006) but for five
growing seasons from 2003/04 to 2007/08. Lucerne produced the greatest liveweight,
33-42% higher than grass-based pastures in 2003/04, 2004/05 and 2006/07 with a lower
number of grazing days. Lucerne had an average of 1620 grazing days/ha 1 over the five
growing seasons compared with 1890 for Rg/Wc and 1266 for Cf/Sc. Lucerne maintained
high liveweight production with fewer grazing days due to superior average daily growth
rates. In spring, hoggets on lucerne pastures averaged liveweight gains of 250 g/hd/d
compared with 195 g/hd/d for Rg/Wc. The same pattern was observed during summer
with lambs grazed on lucerne averaging 160 g/hd/d compared with 65 g/hd/d for grass
with white clover. This highlights the advantage of lucerne to maximise spring and
1
The units reported in this publication (2008b) were later found to have calculation error and were GD/plot
not GD/ha
7
summer liveweight gains compared with ryegrass/white clover when there is a soil
moisture limitation. Lucerne treatments produced greater than 60% more dry matter
annually in 2004/05 and 2006/07 than any grass based treatment with yields of 18500
and 17400 kg DM/ha, respectively.
The daily liveweight gains of lambs grazing lucerne during summer months are
comparable with those recorded by Douglas et al. (1995) who had weaned lambs grazing
lucerne pastures. Daily growth rates on lucerne were 186 g/hd/d for males and 178
g/hd/d for females. They also measured the performance of ewes with lambs at foot on
lucerne pastures. Liveweight gains for Romney ewes were 59 g/hd/d and 263 g/hd/d for
lambs in their experiment based in Manawatu.
On a seasonal basis, Mills et al. (2008b) showed spring was the most reliable and
productive season. Over five years, spring accounted for an average of 64% of the annual
liveweight production and 40-63% of the total dry matter yield. Summer liveweight
production for the grass based pastures was highly variable due to unreliable summer
rainfall, ranging from 65-185 mm for the months December-February. When summer
rainfall was below average, liveweight production accounted for 15-18% of annual
production. However when rainfall was above average, liveweight production accounted
for >30%. Animal production in autumn was influenced by summer rainfall. Liveweight
production after moist summers represented 2-3% of the total annual production
compared with 16-26% after a dry summer. Spring is the most important time to
maximise pasture growth as soils are moist, have low evapotranspiration rates and
therefore water stress is usually low which provides reliable pasture growth (Brown et al.,
2006). High animal growth rates are required during spring, particularly for lamb
finishing, to allow destocking before summer dry. Results from the ‘Maxclover’
experiment and others (Douglas, 1986) confirm lucerne as a reliable and important
source of feed in summer dry environments.
2.3 Animal requirements
The production of a high dry matter yield is of little value if the feed does not meet
animal requirements. The total energy requirement for a given animal is the sum of its
8
requirements for maintenance, liveweight gain, pregnancy and lactation. This can be
expressed by its metabolisable energy (ME) and daily requirement (Nicol and Brookes,
2007). Maintenance is defined as the amount of energy required for an animal to
maintain a constant body weight and to sustain basic processes required for life.
Maintenance requirements are affected by various factors such as age, sex, topography,
climate and physiological state. The requirement is based on bodyweight and also
topography. For example, a 60 kg ewe on flat land requires 9 MJ ME/day for
maintenance, but foraging on hard hills requires 11 MJ ME/day (Appendix 1).
The requirement of ewes during pregnancy differs according to the time of pregnancy
and lamb birth weight. For example, the 60 kg ewe two weeks away from lambing a 4 kg
lamb, will require 4 MJ ME/day in addition to maintenance. If this ewe is bearing twin
lambs, the value needs to be doubled. Therefore, she will require 8 MJ ME/day above
maintenance, giving a total daily requirement of 17 MJ ME/day. Pregnancy requirements
can also be calculated on a flock basis. For example, if a flock scanned 140% and had an
estimated birth weight of 5 kg two weeks out from lambing, requirements would be 1.4 x
5 = 7 MJ ME/day in addition to the maintenance requirement.
The requirements of ewes during lactation also differ depending on the lamb weaning
weight. A ewe bearing twin lambs with a weaning weight of 35 kg requires 29 MJ ME/day
(14.5 MJ ME/day x 2) in addition to maintenance two weeks after lambing. By 12 weeks
after lambing, the requirement is an additional 52 MJ ME/day (26 MJ ME/day x 2). The
ability to calculate the animal requirements at various physiological stages allows feed
budgeting to determine whether the available feed supply is above, below or equal to
animal demand. It also allows forward planning if supplementary feed needs to be
brought in and reduces risk. To meet these animal requirements the quality of feed is
important, particularly during periods of high animal demand, such as during lactation.
2.4 Quality
Thus, the quality of pastures has a direct impact on the animal performance. Mills and
Moot (2010) reported the annual ME energy yields of the six ‘Maxclover’ dryland
pastures six and seven years after establishment. Lucerne monocultures produced ∼134
9
GJ ME/ha/yr which was higher (P<0.001) than all pastures, particularly ryegrass/white
clover producing ∼18 GJ ME/ha/yr. The high quality of lucerne produced superior
liveweight gains compared with ryegrass/white clover in a dryland situation. The ME
content of lucerne averaged ∼11 MJ/kg DM compared with cocksfoot with an ME of
∼11.2 MJ/kg DM.
Annual ME intake from lucerne was reported by Brown et al. (2006), with greater intake
from lucerne (142-261 GJ/ha) than chicory (99-169 GJ/ha) and red clover (74-218 GJ/ha)
for all five regrowth seasons. This explains the superior growth rates on lucerne pastures
due to higher quality feed compared with chicory and red clover, as a result of greater
ME content and dry matter yields.
In addition to superior ME production in Mills and Moot (2010), lucerne also produced
more (P<0.001) N annually than any other pasture with 510 kg N/ha/yr. Ryegrass/white
clover pastures only produced 151 kg N/ha/yr. Lucerne contained an average of 3.9% N
over the two years compared with 3.0-3.5% N for cocksfoot pastures. Increasing N
content up to 5.5% increased the photosynthetic rate in cocksfoot pastures, resulting in
greater dry matter yields (Peri et al., 2002). This shows that lucerne is a higher quality
pasture due to greater ME yields and N content which support superior animal
production.
Tonmukayakul et al. (2009) also reported on the ‘Maxclover’ experiment for the 2008/09
growing season. Annually, lucerne produced the highest N yield of 471 kg/ha of the six
dryland pastures. Cocksfoot/subterranean clover produced 188 kg N/ha which was the
next highest, with subterranean clover contributing 51 kg N/ha with the balance from
cocksfoot. The lucerne contained 3.9% N compared with 3.5 and 4.3% for cocksfoot and
subterranean clover, respectively. This confirms that lucerne should be sown where
possible in a dryland farming system due to higher ME and N production resulting in
superior liveweight production and liveweight gains.
Brown and Moot (2004) investigated the quality of lucerne, red clover and chicory over
six years under irrigation. The quality of palatable, unpalatable, weed and post-grazing
10
residuals were determined. In the palatable fraction, the crude protein (CP) was highest
for lucerne (0.29 g/g DM) followed by red clover (0.25 g/g DM) and then chicory (Table
2.1). However, it is likely that a crude protein content 0.29 g/g DM is above animal
requirements (Waghorn and Barry, 1987). The palatable fractions had substantially higher
CP values than the unpalatable fractions for lucerne (0.12 g/g DM) and chicory (0.08 g/g
DM). The CP value of the post-grazing residual was similar to that of the unpalatable
fraction for lucerne and chicory. This shows that lucerne has more available protein than
chicory and red clover, and this probably supported the reported superior liveweight
gains (Brown et al., 2006).
Brown et al. (2006) also reported annual CP yield for lucerne (3.3-6.3 t/ha) was 1.0-3.6
t/ha greater than for chicory and red clover over all five regrowth cycles. There was a
decline in the annual CP intake over the duration of the experiment with lucerne
decreasing from 6.3 t/ha to 3.4 t/ha.
Table 2.1 Crude protein (g/g DM) and ME (MJ/kg DM) contents of herbage fractions
(palatable, unpalatable and weed) and post-grazing residual of chicory, lucerne
and red clover swards grown over five years. Values in parenthesis are
standard errors for each mean calculated from five year’s data. Adapted from
Brown et al. (2004).
Species
Chicory
Lucerne
Red clover
ME
Chicory
Lucerne
Red clover
* Weed fractions not analysed
CP
Palatable
0.18 (0.011)
0.29 (0.008)
0.25 (0.011)
11.3 (0.20)
11.6 (0.13)
10.9 (0.21)
Unpalatable
0.08 (0.013)
0.12 (0.008)
9.4 (1.39)
7.8 (0.42)
Weed
0.25 (0.011)
*
*
11.4 (0.33)
*
*
Residual
0.10 (0.010)
0.12 (0.013)
0.20 (0.009)
8.6 (0.67)
6.8 (0.55)
10.0 (0.09)
The impact of increasing standing herbage yield from 700 kg DM/ha to 4300 kg DM/ha on
the crude protein content of lucerne was reported by Brown and Moot (2004). Crude
protein content showed an exponential decrease from 0.35 g/g DM to 0.27 g/g DM
11
where it remained stable. While the unpalatable fraction remained constant at 0.11 g/g
DM over the same range. Increasing herbage height had no effect on ME content of both
palatable (11.9 MJ/kg DM) and unpalatable (7.9 MJ/kg DM) components which remained
constant. These data show how important grazing management is in maintaining the CP
content of lucerne stands. However, Waghorn and Barry (1987) reported that CP content
of 0.27 g/g DM is likely to be above animal requirements. The ME content remained
stable with increased herbage, which shows that lucerne quality was not affected by the
height of the stand. However, with increased height the fraction of palatable material in
the stand decreased. Therefore, a lower utilization rate occurred rather than any
influence on quality.
2.5 Grazing selection
Grazing selection of livestock can influence the quality of pastures. Keogh (1986) reported
that livestock selected urine patches in preference to inter-urine patches, due to higher
intensity and more frequent defoliation of urine patches. Edwards et al. (1993)
investigated the effect on N applications of grazing selection. Calcium ammonium nitrate
(26,0,0,0) was applied at zero or 300 kg N/ha to old runout lucerne pastures overdrilled
with seven grass species. Plots were grazed by ewes with lambs at foot. Grass height of
cocksfoot plots with N decreased from 120 mm to 60 mm in the first day of grazing while
plots with zero N applied remained constant at 60 mm. Cocksfoot with 300 kg N/ha had a
N content of 5.10% compared with 3.66% for cocksfoot with zero N applied. This
indicated that sheep had a strong preference for plant species higher in N. Grazing
selection of livestock has the potential to change the dynamics of a pasture from legume
dominant to grass dominant due to the strong preference for legume species. Livestock
can also select between the parts of a plant species. Arnold (1960) reported that sheep
continuously selected leaf in preference to stem. In his study lucerne stands contained
42% stem and 48% leaf prior to grazing with merino wethers. Four days after grazing,
stem content had increased to 88% and leaf content had declined to 12%. This indicates
that sheep selected a high quality diet from what is on offer. Therefore, pre-graze
nutritive analysis may not be indicative of what was consumed. Post-graze nutritive
analysis would allow determination of the exact quality of the diet selected by livestock.
12
2.6 Annual dry matter production
Cocksfoot is commonly sown as a companion grass in lucerne stands due to its drought
tolerance. Cullen (1965) compared the growth of pure lucerne with lucerne/cocksfoot
and lucerne/prairie grass mixes over three years. In the first year, lucerne only produced
6250 kg DM/ha compared with 9580 and 12040 kg DM/ha for lucerne/cocksfoot and
lucerne/prairie grass mixes, respectively. There was a significant difference between the
yields of all three pastures. In the second and third years yield differences were smaller.
Lucerne/prairie grass produced the lowest yields for 1958-59 (6840 kg DM/ha) and 19591960 (11760 kg DM/ha) compared with lucerne and lucerne/cocksfoot. This indicated
that the persistence of prairie grass in the mixed lucerne stand was low.
Douglas and Kinder (1973) compared the growth of pure lucerne stands with
lucerne/cocksfoot pastures with two sowing methods. Cocksfoot was either sown in
alternate rows to lucerne, or mixed with the lucerne. Lucerne produced higher yields
than lucerne/cocksfoot mixes in three out of the four years. Cocksfoot sown in a mixture
produced similar yields to pure lucerne stands, with no significant difference between the
two over the four years. Cocksfoot sown in alternate rows with lucerne produced
significantly less than mixed cocksfoot pastures in 1965-66 and 1967-68. The yields for all
three treatments were very low for 1966-67. The low production was attributed to a very
dry spring in 1966, followed by a summer drought. The location of the experiment would
also have had an effect on the production. The experiment was located at Tara Hills High
Country Research Station which has a semi-arid environment with an average rainfall of
526 mm. An average of 160 mm falls during summer therefore the effectiveness is
decreased by high (>30°C) temperatures and high (>5 mm) evapotranspiration rates.
The yield results reported by Douglas and Kinder (1973) contrast those of Cullen (1965).
Cullen (1965) concluded that lucerne/cocksfoot pastures sown in mixed rows proved
superior to pure lucerne due to increased yields in the first year, effective weed control
and improved growth. While Douglas and Kinder (1973) concluded that lucerne sown
alone was superior to lucerne/grass mixes sown in alternate rows, but similar to
lucerne/cocksfoot sown as a mixture. The differences could also be explained by the
location of the experiments.
13
Prairie grass is one of the most common pasture species sown with lucerne. Fraser (1982)
investigated the performance of lucerne and lucerne/prairie grass mixes in Canterbury. In
the first year, there was no significant difference between the yields of pure lucerne and
lucerne/prairie grass stands. In the second year lucerne/prairie grass mixes yielded
significantly more than pure lucerne stands. This can be explained by above average
rainfall (154 mm), and low air temperature in spring 1977. This resulted in rapid grass
growth until late spring. The trial ended after only two years due to a pea aphid
(Acyrthosiphon pisum Harris) attack on the lucerne followed by a wet winter which
depleted the lucerne plant populations. The annual production of 22370 kg DM/ha in
1977-78 for lucerne/prairie grass mixes was 86% higher than the 12040 kg DM/ha
recorded by Cullen (1965). This yield difference was probably due to the location of the
experiments, Cullen (1965) was located at Invermay Research Station which has a colder
climate compared with Canterbury.
Baars and Cranston (1978) compared the annual production of lucerne with
lucerne/prairie grass mixes at three sowing rates – 4, 13 and 22 kg/ha (Table 2.2). In the
first year, lucerne/prairie grass sown at 22 kg/ha produced 85% more herbage than pure
lucerne. However, this was not sustained during the second and third years, with no
significant difference between yields across all treatments. Yields for 1974-75 were low as
the year began on 19 September 1974 when the plots were sown and finished on the 18
June 1975. Sowing rate had no effect on yield which was consistent with Cullen (1965)
who sowed prairie grass at 2.7 kg/ha and recorded yields of 6800 – 12000 kg DM/ha.
Table 2.2 Annual herbage yields of lucerne and lucerne/prairie grass with three sowing
rates over three years. Adapted from Baars and Cranston (1978).
Yield (kg DM/ha)
Treatment
Sowing rate (kg/ha)
1974/75
1975/76
1976/77
Lucerne
2420 b
11090
10030
Lucerne/prairie grass
4
3900 a
11340
10510
Lucerne/prairie grass
13
4640 a
11900
10190
Lucerne/prairie grass
22
4490 a
10800
9800
Values with subscript letters in common are not significantly different at α = 0.05.
14
McKenzie et al. (1990) investigated the productivity and water use of lucerne and
lucerne/grass mixes in Canterbury. Pure lucerne yielded 12700 kg DM/ha which was
significantly higher than the 10400 kg DM/ha for lucerne/prairie grass. Sowing of grasses
provided no significant yield advantage which was consistent with Douglas and Kinder
(1973), Fraser (1982) and Vartha (1973). McKenzie et al. (1990) concluded that
overdrilling a winter active grass species into mature lucerne stands may be the best way
to establish lucerne/grass mixes.
2.7 Botanical composition
One of the challenges of lucerne/grass mixes is maintaining a balance of both species.
The botanical composition of pure lucerne, lucerne/prairie grass and lucerne/cocksfoot
mixes was investigated by Cullen (1965) over three years. In the first year, lucerne
content of all pastures was low ranging from 53% in the pure lucerne down to 7% in the
lucerne/prairie grass. Sowing of perennial grass species with lucerne reduced the weed
content. In the first year for lucerne pastures, 47% of the annual yield was from unsown
species compared with 25% for both lucerne/cocksfoot and lucerne/prairie grass. In the
second year, the lucerne content of all pastures had increased and by the third year, all
pastures were lucerne dominant. Cocksfoot made up 41% of the lucerne/cocksfoot
pasture in the third year, despite its aggressive growth pattern. The defoliation interval
was deemed the key factor in determining whether the sward was cocksfoot or lucerne
dominant. Frequent defoliation favoured cocksfoot growth, while infrequent defoliation
favoured lucerne growth. However, no specific details were provided in regards to the
duration of frequent and infrequent defoliation periods. Prairie grass was the least
persistent over the experimental period, only contributing to 12% of the annual yield in
the third year.
Marsh and Brunswick (1977) investigated the effect of stocking rate on the botanical
composition of lucerne and lucerne/prairie grass mixes (Figure 2.3). Three stocking rates
were implemented low (5.0 cattle/ha), medium (6.67 cattle/ha) and high (10.0 cattle/ha).
Lucerne/prairie grass mixes had higher dead material than pure lucerne. This could have
15
been due to complete canopy cover resulting in shading of the lower leaves, preventing
photosynthesis. The dead material in both pasture types decreased with stocking rate.
This was due to higher stocking rates having lower post-graze residuals. Lucerne/prairie
grass at a low stocking rate had a mean post-graze residual of 1815 kg DM/ha over four
grazing rotations compared with 1270 kg DM/ha and 435 kg DM/ha for medium and high
stocking rates, respectively. The pattern was the same for pure lucerne stands. The
amount of prairie grass present declined with stocking rate, indicating that it was not
suited to heavy grazing. The weed content was lower in lucerne/prairie grass mixes with
medium and high stocking rates compared with pure lucerne. This was consistent with
findings by Cullen (1965), however, it is not known if the difference in weed content was
significantly different between the two pasture types.
Figure 2.3 Botanical composition of lucerne and lucerne/prairie grass pastures under low,
medium and high stocking rates. Where LSR denotes 5.0 cattle/ha, MSR
denotes 6.67 cattle/ha and HSR denotes 10.0 cattle/ha. From Marsh and
Brunswick (1977).
16
2.8 Seasonal production
Vartha (1973) investigated the seasonal production of Medicago glutinosa M. Beib, M.
sativa, lucerne/ryegrass and lucerne/cocksfoot pastures. The seasonal production of the
lucerne/grass mixes was compared under two grazing intervals, infrequent (early
flowering) and frequent. M. glutinosa was not significantly higher producing than M.
sativa over the experimental duration. Seasonal production of lucerne/ryegrass and
lucerne/cocksfoot was different (P<0.05) during the entire experimental period except for
spring and summer in 1966-67. No comparison was made of lucerne/grass mixes with
pure lucerne stands. Both lucerne/ryegrass and lucerne/cocksfoot mixes produced the
most dry matter during spring, over the three years. Lucerne/cocksfoot mixes produced
72% more than lucerne/ryegrass pastures during autumn-winter of 1968. There were
significant differences in yield for all seasons except spring and summer of 1966-67. One
pasture was not superior over the other in regards to annual production.
Lucerne/cocksfoot did produce more during summer while lucerne/ryegrass produced
more during spring which supports their growth patterns and drought tolerance. Grazing
at early flowering of the lucerne produced more (P<0.01) dry matter than frequent
grazing across all seasons, except autumn-winter 1967 where frequent grazing produced
79% more. The biggest difference in seasonal yield occurred in spring 1967 where grazing
at early flowering produced 90% more dry matter than more frequent grazing. The
duration of frequent and infrequent grazing periods was not defined, however grazing
durations lasted between 38 and 70 days.
The results from the experiment were presented in a confusing manner, giving the
impression that M. sativa and M. glutinosa pastures were sown as pure swards. After
reviewing the methodology, both lucerne species were overdrilled with ryegrass or
cocksfoot once established. Therefore, no comparison of the performance of pure
lucerne with lucerne/grass mixes can be made. There was a difference (P<0.001) between
grazing frequencies, but grazing interval for each frequency was not specified. The
methodology stated the grazing intervals ranged from 38 to 70 days for both frequencies
but did not give exact intervals.
17
The seasonal production of lucerne and lucerne/prairie grass mixes was investigated by
Fraser (1982). Lucerne pastures produced 48% of their annual yield in summer of both
years. Lucerne/prairie grass mixes produced 44% of their annual yield in summer of the
first year compared with 44% of the annual yield during spring of the second year. There
were differences (P<0.05) in seasonal yield between lucerne and lucerne/prairie grass
mixes. Lucerne/prairie grass produced 3530 kg DM/ha and 1420 kg DM/ha in autumn
1977 and 1978, respectively. This was more (P<0.05) than the 2860 and 820 kg DM/ha
produced by lucerne monocultures during the same time.
Autumn droughts occurred in both years of the trial, resulting in lower than expected
yields. In spring 1977, rainfall was above average and temperatures remained low which
favoured rapid grass growth into late spring. This was reflected with lucerne/prairie grass
pastures producing 10010 kg DM/ha for spring, which was more (P<0.05) than the 8560
kg DM/ha produced by lucerne monocultures.
The seasonal production of lucerne was also compared with lucerne/prairie grass mixes
by McKenzie et al. (1990). Pure lucerne stands were superior to lucerne/grass mixes
(Table 2.3). Lucerne stands produced 22% more dry matter annually than lucerne/prairie
grass mixes. However, lucerne only yielded significantly more than lucerne/prairie grass
in March 1989. The seasonal yields were not typical of lucerne/grass mixes at Lincoln. The
spring and summer yields were lower and winter yields were higher than previous
experiments (Vartha, 1973; Fraser, 1982). This could be attributed to the unusually dry
spring and summer with only 270 mm rainfall from June to March 1988 which was less
than half of the long term mean, followed by a warm, moist winter (McKenzie et al.,
1990). The temporal supply of pasture is a key factor in matching feed supply with animal
demand, particularly in dryland systems where summer dry is common.
Table 2.3 Seasonal production of lucerne and lucerne prairie grass pastures. Adapted
from McKenzie et al. 1990.
Yield (kg DM/ha
Species
Nov-88
Jan-89 Mar-89 May-89 Sep-89 Total
Lucerne
2400
1900
3200
2100
3100
12700
Lucerne/prairie grass
2000
1400
2200
1700
2900
10400
LSD
480
500
910
450
440
1600
18
Water use
The dry matter production of pastures is a result of the water extracted and the efficiency
at which this water is then used. McKenzie et al. (1990) investigated the water use of
lucerne and lucerne/prairie grass pastures in Canterbury. The total water used did not
differ between pasture types. Pure lucerne used 384 mm compared with 376 mm used by
lucerne/prairie grass. Lucerne had a mean water use efficiency (WUE) of 25 kg
DM/ha/mm which was significantly more (P<0.05) than 20 kg DM/ha/mm for
lucerne/prairie grass for the period from November 1988 to March 1989. After this
period WUE did not differ (P>0.05) between pasture type and ranged from 22 – 30 kg
DM/ha/mm. This was lower than the WUE of 40 kg DM/ha/mm for dryland lucerne
grown on a Wakanui silt loam (Moot et al., 2008). However, McKenzie et al. (1990) stated
that the water use of lucerne could have been underestimated. Lucerne roots were found
down to shingle, but neutron probe access tubes could not be installed that far down.
Therefore, the lucerne could have been using water from between the shingle particles
that was not accounted for. The lower WUE for lucerne/prairie grass could be explained
by the prairie grass being nitrogen deficient, therefore photosynthesis was less efficient.
McKenzie et al. (1990) also compared the rooting depth of lucerne and lucerne prairie
grass pastures. Pure lucerne roots penetrated deeper into the soil profile than
lucerne/prairie grass roots. In the lucerne/prairie grass pastures, lucerne root mass
declined sharply in the top 20 cm of the soil compared with pure lucerne stands. There
were few lucerne roots below 50 cm. This was attributed to increased interspecific
competition in the top layers of the soil profile between lucerne and prairie grass. The
inclusion of grass species at establishment appeared to decrease the ability of lucerne
roots to penetrate the soil profile. This resulted in less water extraction by the lucerne
roots in the lucerne/grass mix and decreased production during drought. One way to
overcome this issue could be to overdrill grass species once the lucerne has established.
Lucerne has the advantage in its ability to extract water from the soil profile, and also
being able to use it more efficiently for herbage yield. It is able to access water deeper in
the soil profile due to its long taproot and can use this water efficiently due to being high
in N which results in higher photosynthetic rates (Peri et al., 2002).
19
2.9 Conclusions
•
Animal liveweight production is superior on lucerne monocultures, there was no
published literature found on liveweight production on lucerne/grass mixes.
•
Animal liveweight gain is influenced by quality which is directly affected by grazing
selection.
•
The productive performance of pure lucerne compared with lucerne/grass mixes
was variable, depending on the location and environmental conditions. Generally,
lucerne/grass mixes were not superior to pure lucerne stands in terms of annual
production.
•
Of the literature reviewed, lucerne/grass mixes tended to remain lucerne
dominant.
•
Inclusion of a grass species with lucerne decreased the weed invasion.
•
Increased stocking rate on both lucerne and lucerne/grass pastures, decreased
the dead material. Prairie grass yield declined with increased stocking rate, due to
its intolerance of heavy grazing.
•
Lucerne/grass mixes produced more annually with infrequent grazing intervals
compared with frequent grazing intervals.
•
Inclusion of grass in a lucerne stand increased the winter and autumn production,
providing forage when lucerne is dormant.
•
Lucerne/grass mixes have lower water use and WUE than pure lucerne stands due
to interspecific competition for soil moisture. There is little New Zealand literature
on the water use of lucerne/grass mixes.
20
3
MATERIALS AND METHODS
3.1 Experimental site
This experiment is located in paddocks C6E, C7W and C7E in the Cemetery Block, Ashley
Dene Research Farm, Canterbury, New Zealand (43°65’ S, 172°32’ E. 39 m a.s.l.). There
are three soil types in the experimental paddocks, Lismore stony silt loam, Lowcliff stony
silt loam and Ashley Dene deep fine sandy loam (Appendix 2). Lismore stony silt loams
have excessive drainage with a water holding capacity (WHC) of 70-100 mm per metre of
soil but only 450-750 mm depth until stones are present in the profile. Lowcliff stony silt
loam soils are imperfectly drained soils with a WHC of 100-120 mm per metre of soil and
450 - 900 mm depth to stones. In contrast to this, Ashley Dene deep fine sandy loams are
moderately well drained soils with 100-160 mm per metre of soil WHC and greater than
900 mm to stones in the soil profile (McLenaghan and Webb, 2012).
3.2 Experimental area
C6E was sown in kale (Brasscia oleracea ssp. acephala) in December 2010 prior to
conventional cultivation. Lucerne, cocksfoot and brome were sown in November 2011.
C7W has been in ‘Kaituna’ lucerne since October 2006. C7E was sown in rape (B. napus
ssp. oleifera) before being sown in lucerne, cocksfoot and brome after conventional
cultivation in November 2011.
Two cultivars of each species were sown; ‘Safin’ and ‘Vision’ cocksfoot, ‘Bareno’ grazing
brome and ‘Atom’ prairie grass and ‘Stamina 5’ or ‘Kaituna’ lucerne. Subdivision fencing
was completed in July/August 2011 where C6E was divided into paddocks 1-6, C7W into
7-12 and C7E into 13-18 (Plate 1). Details of sowing dates, sowing rates and method of
sowing are given in Table 3.1. Grass species had to be resown in February 2012 due to
poor establishment which was probably a result of the seed being sown too deep. Grass
seed for reseeding was broadcast using a Fiona drill. Lucerne in paddock C6E was also
resown due to poor establishment, and was drilled using a Duncan drill.
21
Table 3.1 Species cultivar, sowing date, rate and drill type for paddocks
C7E at Ashley Dene, Canterbury.
Paddock
Sowing date Species
Cultivar
Sowing rate
(kg ha-1)
C6E
18/11/2011 Cocksfoot
‘Safin’
2
‘Vision’
2
Brome
‘Bareno’
10
‘Atom’
10
19/11/2011 Lucerne
‘Stamina 5’
8
20/02/2012 Cocksfoot
‘Safin’
3
‘Vision’
3
29/02/2012 Brome
‘Bareno’
10
‘Atom’
9
C7W
13/10/2006 Lucerne
‘Kaituna’
10
20/02/2012 Cocksfoot
‘Safin’
3
‘Vision’
3
Brome
‘Bareno’
10
‘Atom’
9
C6E
18/11/2011 Cocksfoot
‘Safin’
2
‘Vision’
2
Brome
‘Bareno’
10
‘Atom’
10
19/11/2011 Lucerne
‘Stamina 5’
8
13/12/2011 Lucerne
‘Stamina 5’
8
20/02/2012 Cocksfoot
‘Safin’
3
‘Vision’
3
29/02/2012 Brome
‘Bareno’
10
‘Atom’
9
C6E, C7W and
Drill type
Triple disc
Triple disc
Triple disc
Triple disc
Duncan
Fiona
Fiona
Fiona
Fiona
commercial
Fiona
Fiona
Fiona
Fiona
Triple disc
Triple disc
Triple disc
Triple disc
Triple disc
Duncan
Fiona
Fiona
Triple disc
Triple disc
3.3 Experimental design
The experiment covers a total of 17.7 ha consisting of 18 paddocks with three cultivar
replicates or six species replicates. A large mob of 650 ewes grazed all three paddocks
starting on the 28 June 2012 in C6E and finishing on the 4 July 2012 in C7E. For this
experiment, spring grazing with ewes and lambs commenced on 5 September 2012.
22
Plate 1 Map of experimental design showing paddocks C6E, C7W and C7E and plots 1-18
at Ashley Dene, Canterbury. The total experimental area is 17.7 ha.
3.4 Soil fertility
Table 3.2 shows the results from soil samples taken during the months of May and June
2011.
Table 3.2 Soil test results from May/June 2011 for paddocks C6E, C7W and C7E, Ashley
Dene, Canterbury, New Zealand.
Soil test results
Optimum
C6E
C7W
C7E
pH
6-6.5
5.7
5.5
5.8
Olsen P
20-30
14
23
19
K (me/100 g)
6-12
0.39
0.40
1.19
Ca (me/100 g)
0.5-12
6.6
6.5
8.2
Mg (me/100 g)
0.8-3.0
0.57
0.67
0.70
Na (me/100 g)
0.1-0.5
0.12
0.12
0.13
CEC (me/100 g)
20-25
15
14
15
Total base saturation
55-75
53
55
68
Sulphate sulphur (mg/kg)
10-20
3
5
18
23
3.5 Fertiliser
Based on the soils test 2 t/ha of lime was applied over all paddocks in September 2011. In
September 2012 Sulphur Super 15 (0,9,0,15) was applied to C6E at a rate of 250 kg ha-1
and 350 kg ha-1 in C7W and C7E.
3.6 Meteorological data
Mean monthly air temperatures were recorded approximately 14 km away from the
experimental site at Broadfields weather station (43°62’S, 172°47’E). Monthly rainfall
data were recorded at the Ashley Dene weather station located within the Cemetery
Block, paddock C2 (43°65’S, 172°32’E). The data are shown in Figure 3.1 along with the
long term averages for air temperature (1975-2010) and rainfall (1980-2009). Monthly
rainfall was variable between 50 and 100 mm until June 2013 when ∼ 170 mm fell. This
had little impact on the animal or dry matter results because it was in the last month of
measurements when plots were destocked. The temperature data were within the
normal range, being the highest in January.
24
Figure 3.1 Mean monthly rainfall (a) and air temperature (b) for the 2012/2013 growing
season with long term means for the period 1975-2010 (air temperature) and
1980-2009 (rainfall). Air temperature data were obtained from Broadfields
meteorological station (43°62’S, 172°47’E). Rainfall data were obtained from
Ashley Dene weather station (43°65’S,172°35’E).
25
3.7 Soil water budget
3.7.1 Potential soil water deficit
Potential soil moisture deficit (PSMD) from the 1 July 2012 to 30 June 2013 is shown in
Figure 3.2. PSMD was set at zero on 1 July 2012 and accumulated from then on using
Equation 1.
Equation 1
Todays PSMD = Yesterdays PSMD + Todays Penman PET – Today s rainfall
Negative PSMD values were not allowed to be returned. Rainfall and Penman potential
evapotranspiration (PET) data were obtained from Broadfields meteorological station
(43°62’S,172°47’E). PSMD increased from zero on 1 July 2012 to a maximum of 595 mm
on 17 May 2013. PSMD is a calculated estimate based on climatic data, and is not an
indication of the actual soil moisture deficit. This is found by examining the soil water
content.
Month
Jun12
0
Aug12
Oct12
Dec12
Feb13
Apr13
Jun13
PSMD (mm)
200
400
600
Figure 3.2 Potential soil moisture deficit (PSMD, mm) between 01/07/2012 and
31/05/2013 for paddocks C6E, C7W and C7E at Ashley Dene, Canterbury, New
Zealand.
26
3.7.2 Soil water content (SWC)
Volumetric soil water content was measured throughout the experiment. A Time Domain
Reflectometer (TDR) was used to take measurements at 0-0.2 m and a neutron probe
(Troxler) was used for measurements every 0.2 m from 0.25-2.25 m. This allowed
calculation of the soil water content (SWC) and plant available water. The drained upper
limited which represents field capacity, was determined as the average of the second and
third highest values obtained throughout the experimental period. The highest value was
not used to prevent using full saturation before drainage as the upper limit. The lower
limit or permanent wilting point was taken as the lowest soil water value obtained during
the experimental period and occurred between March and May 2013 for all plots. The
total available water was then calculated using Equation 2.
Equation 2
Plant available water content = ∑(drained upper limit) – ∑(lower limit)
3.7.3 Water use efficiency
The amount of water used by plants was calculated to determine the water use efficiency
(WUE) of the pastures and was calculated using Equations 3 and 4 (Sim et al., 2012).
Equation 3
𝑾𝑼 = 𝑷𝑹 − (𝑺𝑾𝑪𝑬 − 𝑺𝑾𝑪𝑺 )
Where PR is the sum of rainfall for the specified period, SWCE is the actual soil water
content of the profile at the start of the measurement period and SWCS is the actual soil
water content at the end of the measurement period.
The daily Penman potential evapotranspiration (PETdaily) for duration of the measurement
period was then used to calculated daily water use (WUdaily).
Equation 4
𝑾𝑼
𝑾𝑼𝒅𝒂𝒊𝒍𝒚 = �𝑷𝑬𝑻� ∗ 𝑷𝑬𝑻𝒅𝒂𝒊𝒍𝒚
27
3.8 Livestock and grazing management
The livestock used for this experiment were sourced from the Lincoln University
Coopworth flock. A summary of the rotation dates and stock classes are given in Table
3.3. Detailed stock movements on a treatment and plot basis are given in Appendix 3. On
5 September 2012, 41 ewes bearing twin lambs were weighed and began grazing one plot
for each treatment at approximately 3.7 SU/ha (Table 3.4). Stocking rates were calculated
as the mean of the 6 plots for each treatment. A stock unit is defined as one breeding
ewe that weighs 55 kg and bears one lamb, consuming 550 kg DM per year. This also
includes the feed consumed by the lamb until weaning at 3.5 months old (Trafford and
Trafford, 2011). The stocking rate of each treatment was determined visually estimating
the feed available and completing a feed budget to determine how many animals could
be grazed and how long for. The stocking rate was increased as stock became available to
15 SU/ha to cope with the rapid spring pasture growth. Stock were weighed full,
approximately monthly. The ewes and lambs completed two rotations of the
experimental site before being removed for weaning on 28 November 2012. Stock were
shifted once the desired grazing residual was meet. Three plots were grazed at once, with
one from each treatment.
Lambs were weaned then returned to the experimental site at 8 SU/ha on the 28
November 2012. The stocking rate decreased from then on as lambs were removed for
slaughter when they reached killable weights of 34 kg liveweight. After two rotations, all
lambs were removed and weighed on the 25 January 2013. A cleanup graze was then
completed across all pastures with ewes at 30 SU/ha. Ewes were removed from the trial
by the 4 February 2013. Dry conditions meant there was little growth during February so
a second clean up graze was completed in March 2013 using ram hoggets to again graze
all plots at 30 SU/ha.
A final live weight measurement period commenced on 15 May 2013 with ewe hoggets
stocked at 6.5 SU/ha. This was increased to 11 SU/ha based on dry matter data and feed
intake. Stock were removed from the experimental site on the 26 June 2013, fasted
overnight and weighed. A cleanup graze was then completed with ewes at 32 SU/ha
starting on the 1 June 2013 and finishing on the 3 July 2013.
28
Each plot in the experiment is surrounded by permanent netting fences. Small plastic
troughs in each paddock supplied water to stock.
Table 3.3 Summary of stock class, start and end date and plots grazed for each grazing
rotation at Ashley Dene, Canterbury, New Zealand. Where E & L denotes ewes
and lambs, W L denotes weaned lambs, Ram Hgts denotes ram hoggets and
Ewe Hgts denotes ewe hoggets.
Rotation
1
2
3
3a
Cleanup 1
4
5
Plots grazed
1-18
1-18
1-18
1-3
1-18
1-18
1-18
Stock Class
E&L
E&L
WL
WL
Ewes
Ram Hgts
Ewe Hgts
Start date
5/09/2012
24/10/2012
28/11/2012
11/01/2013
22/01/2013
5/03/2013
15/05/2013
End date
24/01/2012
28/11/2012
11/01/2013
19/01/2013
4/02/2013
28/03/2013
26/06/2013
Cleanup 2
1-18
Ewes
1/06/2013
3/07/2013
Table 3.4 Grazing rotation stocking rates (SR), expressed as stock units (SU/ha) for
lucerne monocultures, lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br)
pastures for either production or maintenance liveweight (LWT) from 5/09/12
to 30/06/13 at Ashley Dene, Canterbury, New Zealand.
Grazing
rotation
1
1
1/2
3
3
Cleanup 1
4
5
5
5
Cleanup 2
LWT Period
Production
Production
Production
Production
Production
Maintenance
Maintenance
Production
Production
Production
Maintenance
Date on
5/09/12
14/09
18/09
28/11/12
8/01
22/01
5/03
15/05
30/06
12/06
17/06
Date off
14/09
18/09
28/11
8/01/13
11/01
4/02
28/03
30/05
12/06
26/06
30/06/13
Lucerne SR
(SU/ha)
3.4
8.7
15.7
8.2
4.4
31.8
25.1
7.4
9.4
12.2
34.7
Luc/CF SR
(SU/ha)
3.7
10.1
15.7
7.8
3.9
31.8
21.8
7.3
9.4
12.2
34.7
Luc/Br SR
(SU/ha)
3.7
10.1
15.4
7.3
4.6
31.8
27.9
7.7
9.8
12.5
34.7
29
3.9 Weed control
Nodding thistles (Carduus nutans) were an issue in the older lucerne stands in C7W. The
herbicide Velpar DF (active ingredient Hexazinone 240 g L-1) was sprayed in October 2011
and January 2012 at the recommended rate of 1.2 kg per 300 L water/ha for control of
these. Velpar can suppress lucerne plants, therefore the stand was grazed severely
before spraying to reduce the leaf area of lucerne plants, to minimize herbicide uptake.
3.10 Measurements
A summary of grazing rotations with the stock class grazing and measurements taken is
given in Table 3.5. Rotations were classified as either ‘production’ periods where
liveweight gain was measured or ‘maintenance’ periods where cleanup grazing occurred
and animals were not measured but assumed to maintain a constant weight. Dry matter
yield for rotation ‘Cleanup 2’ was estimated at 1000 kg DM/ha for all plots based on
visual assessment (M. Smith 2013, personal communication). LWT measurement periods
did not always coincide with grazing rotations. This was because grazing rotations one
and two, stock were weighed three times.
Table 3.5 Summary of stock classes and measurements taken for grazing rotations from
1/07/12 to 30/06/13. Where E & L denotes ewes and lambs, W L denotes
weaned lambs, Ram Hgts denotes ram hoggets and Ewe Hgts denotes ewe
hoggets. Measurements taken include dry matter yield (DM), animal liveweight
gain (LWt), botanical composition (BC) and nutritive value (NU) and are
indicated by a ‘Y’. LWT period determines if the rotation was a ‘production’ or
‘maintenance’ period. Liveweight (LWT) rotation is when stock were weighed
relative to grazing rotations.
Grazing
Rotation
1
1
2
3
3a
Cleanup 1
4
5
Cleanup 2
Plots Grazed
1-6, 16-18
1-18
7-15
1-18
1-3
1-18
1-18
1-18
1-18
Stock
Class
E&L
E&L
E&L
WL
WL
Ewes
Ram Hgts
Ewe Hgts
Ewes
Measurements
DM
LWt BC
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
NU
Y
Y
Y
Y
Y
LWT
Rotation
LWT
Period
1
2
3
4
5
Production
Production
Production
Production
Maintenance
Maintenance
Maintenance
Production
Maintenance
30
3.10.1 Live weight measurements
Stock used during the experimental period were weighed using a Tru Test XR3000 system
attached to a Prattley weigh crate. During grazing rotations one, two and three, stock
were weighed on and off in satellite yards without fasting. Stock used grazing rotation
five were fasted overnight before weighing in the main yards. For this experiment ‘spring
grazing ‘was defined as the period of grazing with ewes and lambs from the 5 September
to 28 November 2012, ‘summer grazing’ as the period of grazing with weaned lambs from
28 November 2012 to 11 January 2013 and ‘autumn grazing’ 15 May to 26 June 2013 with
ewe hoggets.
The total number of graze days for the experimental period was derived by multiplying
the number of stock/ha by the duration of grazing. This was then broken down into
‘maintenance’ and ‘production’ periods (Table 3.5).
3.10.2 Dry matter measurements
Automated sward stick readings were taken pre and post grazing to measure sward
height. Within each plot, 50 height measurements were taken. The start and end values
on the clicker were recorded for each plot. These values were calculated by measuring
the distance on the shaft of the stick from the base of the sward to the top of the sward
by moving a slide tube. The distance that the slide tube traveled was calculated for each
plot using Equation 5.
Equation 5
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑙𝑖𝑐𝑘𝑠
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑒𝑎𝑑𝑖𝑛𝑔𝑠
/2
Height measurements were then calibrated with three quadrat cuts to determine the dry
matter yield. A 0.2 m2 quadrat was placed in a location representative of the entire
paddock. The height of the quadrat sample was measured before plant material was cut
with hand shears. Each sample was placed in a paper bag and stored at 4°C until it was
processed. When processed, samples were sorted into their botanical components,
lucerne, grass (cocksfoot or brome), weeds and dead material. Where the sample size
allowed, lucerne stems were cut, separating the top of the plant from the bottom for
comparison of nutritive quality. The samples were then dried in a forced air oven at 60°C
31
for a minimum of 48 hours until they reached a constant weight. Samples were then
weighed using Mettler Toledo PB1502 and Sartorius 3716 electronic scales.
Height and dry matter measurements were used to produce linear regressions of the
relationship as a method of calculating dry matter yields (Figure 3.3). Three destructive
quadrat cuts of high, medium and low heights were taken for each plot. The DM yield was
determined for each of these. The height of each destructive cut was plotted against its
corresponding DM yield and a linear regression was fitted based on the entire season
from these cuts and whether they were pre or post graze. For pre graze yields, spring and
summer regressions were not different (P<0.642) therefore one regression was plotted
for both seasons. A separate regression was fitted for autumn data. For post graze yields,
spring, summer and autumn regressions were not different (P<0.168), therefore one
regression was fitted for all three seasons.
Plate 2 Ewe hoggets grazing lucerne/brome pastures on 27 May 2013 at Ashley Dene,
Canterbury, New Zealand. Reproductive stems are visible, highlighting the
quality decline.
32
Plate 3 Botanical composition of a lucerne/cocksfoot pasture on 16 May 2012 from
Ashley Dene, Canterbury, New Zealand. Botanical components are lucerne,
cocksfoot, weeds and dead material.
33
Total DM Yield (kg DM/ha)
(a)
6000
4000
2000
0
Total Post-graze DM yield (kg DM/ha)
(b)
6000
4000
2000
0
0
20
40
60
Lucerne height (cm)
Figure 3.3 Pre (a) and post-graze (b) linear regressions of lucerne height versus dry matter (DM)
yield for spring (●), summer (○), and autumn (▽) at Ashley Dene, Canterbury. Forms
of the regression were: Spring/summer pre-graze yield = 261±136 + 84±4.2x R2=0.70,
autumn pre-graze yield = 510±130 + 42±6.5x R2=0.43. Post-graze
spring/summer/autumn yield = 198±85.6 + 91.4±4.36x R2=0.64.
3.10.3 Lucerne quality
Samples taken throughout the duration of the experimental period were analysed for
nutritional quality. Pre-graze samples were obtained for grazing rotations one, two, three
and five but post-graze samples for rotation three only. The three destructive cuts were
34
combined, into one sample which was then analysed by infrared spectrometry (NIRS) to
determine the nutritive quality.
3.10.4 Thermal time
Thermal time was calculated to determine the relationship between yield and air
temperature. Temperature data used were from Broadfields meteorological station. In its
simplest form calculation of thermal time requires a daily maximum (Tmax) and daily
minimum (Tmin) temperature and a specified base temperature (Tb) (McKenzie et al.,
1999) Equation 6.
Equation 6
Tt (°Cd) =
𝑻𝒎𝒂𝒙−𝑻𝒎𝒊𝒏
𝟐
− 𝑻𝒃
Two temperature thresholds were tested. The first was a two-stage model with a linear
increase from a specified base temperature (Tb) to an optimum temperature (To) of 30°C
followed by a linear decline to a maximum temperature (Tm) of 40°C (Brown et al., 2005).
The Tb tested ranged from 0-10°C. The second model tested was a broken-stick model
with a linear increase from Tb = 1°C of 0.71°Cd for every 1°C increase in mean air
temperature to an inflection point of 15°C. Above this a linear increase of 1°Cd for every
1°C increase in mean air temperature is observed to To of 30°C, then a linear decrease to
Tm of 40°C (Moot et al., 2001).
The model used was determined by fitting linear regressions to cumulative yield versus
thermal time.
3.10.5 Selection of the base temperature
Figure 3.4 shows Tb and the corresponding coefficient of determination (R2) values for a
two-stage model with Tb 0-10°C and the three-stage model with a Tb of 1°C. The Tb was
selected based on the highest R2 value. The two-stage model with a Tb of 0 had the
highest coefficient of determination value of 0.99 for all pastures.
35
1.000
0.995
0.985
2
R value
0.990
0.980
0.975
0.970
0
2
4
6
8
10
Base temperature (Tb)
Figure 3.4 Base temperatures (Tb) and corresponding R2 values for lucerne monocultures
(●), lucerne/cocksfoot (■) and lucerne/brome (▽) pastures using a twostage model for spring 2012 data. The gray area indicates the R2 values for a
three-stage model with a Tb=1°C.
3.10.6 Statistical analysis
Statistical analyses were carried out in Genstat 15 (Version 15, VSN International Ltd,
Hemel Hempstead, UK). Data for individual plots were tested by one-way analysis of
variance (ANOVA). For analysis of pasture mixes there were 6 replicates of 3 pasture
mixes (d.f.=17). A Fisher’s protected least significant difference (LSD) test was used to
determine differences in annual and rotational yield and botanical composition.
Annual animal liveweight gain was analysed by one-way ANOVA using the plots as
replicates. Rotational liveweight gain was analysed by a t-test due to no mob replicates.
Differences in the standard error of the mean (SEM) between treatments for animal
liveweight gains are due to there being different numbers of stock grazing each
treatment.
36
Thermal time, WUE and height against dry matter were analysed by fitting linear
regressions. Regressions and coefficients of determination (R2) values for spring thermal
time and spring WUE were fitted in Microsoft Excel 2010, then the regression coefficients
were analysed in Genstat by one-way analysis of variance in randomized blocks. No
statistics were run on the phase two regression for thermal time.
37
4
RESULTS
4.1 Animal production
4.1.1 Annual liveweight production
Annual production from the five measured periods totalled 865 kg LWT/ha from lucerne
monocultures and lucerne/cocksfoot pastures which was 16% more (P<0.001) than the
746 kg LWT/ha from the lucerne/brome pasture (Figure 4.1). Spring liveweight
production from ewes and lambs was 570 kg LWT/ha for lucerne monocultures and 584
kg LWT/ha for lucerne/cocksfoot mixes which was higher (P<0.001) than the 537 kg
LWT/ha produced from lucerne/brome mixes. Summer liveweight production from
weaned lambs was highest (P<0.001) in lucerne monocultures and lucerne/cocksfoot
mixes with 252 kg LWT/ha and 236 kg LWT/ha respectively. Lucerne/brome produced
149 kg LWT/ha for the summer period with weaned lambs. Autumn liveweight
production from ewe hoggets was the highest (P<0.001) for lucerne/brome mixes with 59
kg LWT/ha compared with 43 kg LWT/ha for lucerne monocultures and 46 kg LWT/ha for
lucerne/cocksfoot mixes.
38
Accumulated liveweight production (kg/ha)
1200
1000
800
600
400
200
0
Luc
Luc/Brome
Luc/CF
Pasture
Figure 4.1 Annual liveweight production of lucerne monocultures, lucerne/brome
(Luc/Br) and lucerne/cocksfoot (Luc/CF) mixes over five liveweight production
periods from 1/07/2012 to 30/06/2013 at Ashley Dene, Canterbury, New
Zealand. Stacked bars represent spring liveweight gain with ewes and lambs
(■), summer liveweight gain with weaned lambs (▩) and autumn liveweight
gain with ewe hoggets (■). The error bar is SEM for accumulated liveweight
production.
4.1.2 Rotational liveweight production from ewes and lambs
In grazing rotation one and two, 78% of the total liveweight production was from lambs
for lucerne monocultures and lucerne/brome mixes compared with 80% for
lucerne/cocksfoot mixes (Figure 4.2). Ewe liveweight production for grazing rotation one
was 66 kg/ha for lucerne monocultures which was higher (P<0.01) than the 62 kg/ha for
lucerne/grass mixes. In grazing rotation two, lucerne monocultures produced 60 kg
LWT/ha for ewes which was again more (P<0.001) than the lucerne/grass mixes at 57
kg/ha. In grazing rotation one, lucerne/cocksfoot produced 243 kg LWT/ha for lambs
which was higher (P<0.001) than lucerne monocultures and lucerne/brome mixes.
39
Lucerne/cocksfoot was also superior (P<0.001) in grazing rotation two producing 222 kg
LWT/ha.
300
(a)
(b)
(c)
(d)
250
Accumulated liveweight production (kg/ha)
200
150
100
50
0
300
250
200
150
100
50
0
Lucerne
Luc/CF
Pasture
Luc/Br
Lucerne
Luc/CF
Luc/Br
Pasture
Figure 4.2 Spring liveweight production (kg LWT/ha) for rotation one for ewes (a) and
lambs (c) and rotation two ewes (b) and lambs (d) grazing lucerne
monocultures (■), lucerne/cocksfoot (▩) and lucerne/brome (■) mixes at
Ashley Dene, Canterbury, New Zealand. The error bars are SEM for liveweight
production across treatments.
4.1.3 Total graze days
There was no difference (P<0.462) in annual grazing days across treatments. However,
lucerne monocultures and lucerne/cocksfoot mixes had 3830 production graze days
which was higher (P<0.001) than lucerne/brome with 3550 production graze days (Table
4.1). There was no difference in maintenance graze days across treatments (P<0.248).
40
Table 4.1 Total, maintenance and production graze days (GD/ha) for lucerne
monocultures, lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes
from 1/07/12 to 30/06/13 at Ashley Dene, Canterbury, New Zealand.
Production GD are for rotations when liveweight was measured and
maintenance GD are for rotations when liveweight was not measured.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Maintenance GD
711
752
859
784
66.9
0.338
Production GD
1695 a
1708 a
1632 b
1678
24.3
0.021
Annual GD
3060
3150
3064
3071
76.3
0.657
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
4.1.4 Investigating the effect of pasture age
A t-test for lucerne monocultures, showed newly established swards produced 996±43.6
kg LWT/ha which was 65% more (P<0.01) than the 602±45.5 kg LWT/ha from older
overdrilled lucerne stands.
4.1.5 Ewe liveweight
Ewes on lucerne/cocksfoot pastures were the heaviest initially weighing 67 kg compared
with 65 kg for ewes grazing lucerne/brome and 64 kg for lucerne monocultures (Figure
4.3). At the end of the liveweight trial, ewes grazing all trials had gained approximately 11
kg liveweight.
41
Figure 4.3 Change in lactating ewe liveweight over two dry matter rotations from 5/09 to
24/11/12 on lucerne monocultures (●), lucerne/cocksfoot (■) and
lucerne/brome (▽) mixes at Ashley Dene, Canterbury, New Zealand.
4.1.6 Rotational animal liveweight gain
Rotation 1 liveweight gains (g/head/d) of ewes and lambs were similar (P<0.929) on all
pastures (Table 4.2). In rotation two, ewe liveweight gain was 212 g/head/d on lucerne
monocultures which was 25% higher (P<0.015) than lucerne/cocksfoot mixes and 17%
higher (P<0.0.034) than lucerne/brome mixes. Lamb liveweight gains during rotation two
were not different (P<0.906) across all treatments. In rotation three, there was no
difference (P<0.819) in liveweight gain of both ewes and lambs across all treatments.
42
Table 4.2 Liveweight gain (g/head/d) of ewes and lambs grazed on lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes over liveweight
Rotations 1, 2 and 3 from 5/09 to 23/11/12 at Ashley Dene, Canterbury, New
Zealand.
Treatment
Ewes:
Lucerne
Luc/CF
Luc/Br
Lambs:
Lucerne
Luc/CF
Luc/Br
Rotation 1
LWT
(g/head/d)
SEM
Rotation 2
LWT
(g/head/d)
SEM
Rotation 3
LWT
(g/head/d)
SEM
118
166
169
23.9
25.2
26.2
212 a
170 b
181 b
10.2
13.5
10.2
82
106
97
16.3
17.1
19.0
292
301
312
7.9
8.5
8.6
320
326
321
5.3
6.3
6.1
281
291
278
10.3
9.9
9.2
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Differences in SEM’s within a rotation are due to different numbers of stock for
each treatment.
Liveweight gain (g/head/d) of weaned lambs grazing lucerne monocultures over
liveweight Rotation 4 was 243 g/head/d which was 10% higher (P<0.0.29) than lambs
grazing lucerne/cocksfoot mixes and 48% higher (P<0.001) than lambs grazing
lucerne/brome mixes (Table 4.3). Lambs grazing lucerne/cocksfoot mixes gained 221
g/head/d which was 35% higher (P<0.001) than the 164 g/head/d gained by lambs
grazing lucerne/brome mixes.
Table 4.3 Liveweight gain (g/head/d) of weaned lambs grazing lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) pastures over liveweight
Rotation 4 from 28/11/12 to 4/01/13 at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
LWt (g/head/d)
243 a
221 b
164 c
SEM
8.29
5.92
7.92
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Differences in SEM’s within a rotation are due to different numbers of stock for
each treatment.
43
Liveweight gain (g/head/d) of ewe hoggets grazing lucerne/brome pastures in Rotation 5
was 109 g/head/d which was 35% more (P<0.009) than ewe hoggets grazing lucerne
monocultures and 25% more (P<0.025) than ewe hoggets grazing lucerne/cocksfoot
mixes (Table 4.4).
Table 4.4 Liveweight (LWT) gain (g/head/d) of ewe hoggets grazing lucerne
monocultures, lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) pastures
over liveweight Rotation 5 from 15/05 to 18/06/13.
Treatment
Lucerne
Luc/CF
Luc/Br
LWt (g/head/d)
80.6 b
87.2 b
109 a
SEM
8.45
7.07
6.61
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Differences in SEM’s within a rotation are due to different numbers of stock for
each treatment.
4.2 Pasture dry matter yield
4.2.1 Accumulated dry matter yield
There were no differences (P<0.221) in dry matter yield among lucerne monocultures
(12.8 t DM/ha) lucerne/cocksfoot (12.3 t DM/ha) and lucerne/brome pastures (12.7 t
DM/ha) (Figure 4.4). These yield values for 2012/13 are for the liveweight production
periods only, dry matter was not measured during maintenance grazing as indicated by
grey bars. However, dry matter production during this time was estimated at 1000 kg
DM/ha and was included in the accumulated dry matter yield.
44
Figure 4.4 The total accumulated dry matter (DM) yield of lucerne monocultures (●),
lucerne/brome (▽), and lucerne/cocksfoot (■) pastures from 1/07/2012 to
30/06/2013 at Ashley Dene, Canterbury, New Zealand. Grey area indicates the
period when no measurements were taken due to low summer growth.
4.2.2 Mean daily growth rates
For the period of growth from 1/07 to 5/09/12, all pastures (P<0.13) grew at 30±0.68 kg
DM/ha/d (Figure 4.5). During October, there was also no difference (P<0.45) in the
pasture growth rate, which averaged 91±1.5 kg DM/ha/d. During grazing rotation two
from 31/10 to 28/11/12 lucerne monocultures and lucerne/brome mixes grew at 95±1.6
kg DM/ha/d and 98±1.6 kg DM/ha/d, respectively. This was higher (P<0.007) than the
89±1.6 kg DM/ha/d produced by lucerne/cocksfoot mixes. From 25/01 to 5/03/13 the
growth rate of all pastures was 22±0.86 kg DM/ha/d (P<0.86). Growth rates declined
further to 16±0.38 kg DM/ha/d (P<0.71) for all treatments from 8/03 to 27/05/13.
45
Figure 4.5 Mean daily growth rates of lucerne monocultures (●), lucerne/cocksfoot (■)
and lucerne/brome (▽) mixes for regrowth cycles between 1/07/12 and
30/06/13 at Ashley Dene, Canterbury, New Zealand. Error bars are SEM for
each harvest date. Vertical grey bars indicate maintenance grazing periods
where no dry matter measurements were taken.
4.3 Thermal time relationships
The effect of temperature on dry matter production was quantified by calculating
thermal time (Section 3.10.4).
The extrapolated x-axis intercept, which indicates the lag phase, was 340±12.3 °Cd for
lucerne and lucerne/cocksfoot mixes and occurred on the 11/08/2012 (Figure 4.6). The
intercept for lucerne/brome mixes was higher 390±12.3 °Cd (P<0.05) which occurred on
16/08/2012. This represents the lag phase before the linear increase in yield.
The relationship between thermal time and accumulated dry matter was linear during
spring before the rate declined. Growth rates for all pastures averaged 5.5±0.19x –
46
1933±281 kg DM/ha/°Cd during the period when pastures were grazed with ewes and
lambs.
A second regression was fitted from 23/12/12 to 5/06/13 to determine the second slower
phase of growth. Lucerne monocultures grew at a rate of 0.82±0.19 kg DM/ha/°Cd,
lucerne/cocksfoot mixes at 1.07±0.25 kg DM/ha/°Cd and lucerne/brome mixes at
0.68±0.25 kg DM/ha/°Cd.
Regression equations and the coefficients of determination of the regressions of
accumulated dry matter against thermal time for individual plots are given in Appendix 4.
Figure 4.6 Relationship between accumulated dry matter (DM) yield and accumulated
thermal time (°Cd, Tb=0°C) for lucerne monocultures (●), lucerne/cocksfoot (
■) and lucerne/brome (▽) mixes. Forms of the spring regression lines were:
Yield = 5.5±0.19x – 1933±281 (R2=0.99). Thermal time was accumulated using
air temperature. Grey lines extrapolate back to the x-intercept. Full details of
regression in Appendix 4.
47
4.4 Botanical composition
4.4.1 Annual botanical composition
Botanical composition for all pastures was determined pre- (Table 4.5) and post-grazing
(Table 4.6). Prior to grazing, the lucerne component ranged from 80% for lucerne
monocultures which was higher (P<0.001) than 60% for lucerne/cocksfoot and 54% for
lucerne/brome pastures. Weed content was 10% for lucerne monocultures which was
higher (P<0.009) than for the other two treatments. The sown grass component was
similar (P<0.159) for lucerne/cocksfoot and lucerne/brome mixes. Post-grazing, the
lucerne component had decreased to 68% for lucerne monocultures, 48% for
lucerne/cocksfoot and 43% for lucerne/brome mixes. Lucerne monocultures maintained
a higher (P<0.001) proportion of lucerne than the lucerne/grass mixes. The dead material
as a percentage of herbage doubled for all treatments after grazing.
Table 4.5 Annual pre-grazing botanical composition of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
80.6 a
59.9 b
53.5 b
64.6
2.10
<0.001
Sown grass (%)
26.2
30.5
28.4
1.84
0.159
Weed (%)
10.4 a
4.19 b
5.60 b
6.80
1.18
0.009
Dead (%)
8.98
9.69
10.4
9.70
0.97
0.617
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
48
Table 4.6 Annual post-grazing botanical composition of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
68.3 a
47.8 b
42.7 b
52.9
1.75
<0.001
Sown grass (%)
29.2
32.7
30.9
2.09
0.292
Weed (%)
12.4 a
4.07 b
5.99 b
7.50
1.93
0.030
Dead (%)
19.3
18.9
18.6
19.0
1.78
0.963
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
4.4.2 Pasture botanical composition in different grazing rotations
4.4.2.1 Rotation 1
The lucerne component of the pre-graze pastures was 90% in the pure lucerne (P<0.003),
83% in lucerne/cocksfoot and 75% in lucerne/brome (Table 4.7). The weed and dead
components were not different (P<0.674) for all treatments, nor was the sown grass
(P<0.134) for lucerne/grass treatments.
Post-grazing with ewes and lambs, all components remained in similar proportions to
pre-graze values across all treatments (Table 4.8). Lucerne monocultures still comprised a
higher (P<0.024) percentage of lucerne at 90% compared with 80% for lucerne/cocksfoot
and 76% for lucerne/brome mixes. The weed, dead and sown grass components
remained similar (P<0.914) across all treatments.
49
Table 4.7 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for grazing rotation one from 5/09
to 16/10/12 prior to grazing with ewes and lambs at Ashley Dene, Canterbury,
New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
90.1 a
82.5 b
75.1 c
82.5
2.26
0.003
Sown grass (%)
8.90
14.2
11.6
2.09
0.134
Weed (%)
8.80
7.30
9.50
8.60
1.75
0.674
Dead (%)
1.07
1.29
1.21
1.20
0.15
0.597
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
Table 4.8 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for grazing rotation one from
5/09/ to 16/10/12 after grazing with ewes and lambs at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
91.4 a
79.0 b
75.9 b
82.1
3.48
0.024
Sown grass (%)
11.4
11.7
11.5
2.09
0.914
Weed (%)
7.40
8.30
10.8
8.9
2.58
0.638
Dead (%)
1.13
1.38
1.53
1.35
0.31
0.658
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
Ewes and lambs grazing lucerne monocultures consumed 1192 kg DM/ha of lucerne
which was not different to the lucerne component in the lucerne/cocksfoot mixes but
higher (P<0.048) than the component of lucerne/brome mixes (Table 4.9). The sown
grass consumed from lucerne/brome was 191 kg DM/ha and 49 kg DM/ha for
lucerne/cocksfoot pastures (P<0.247). Stock grazing lucerne monocultures consumed 140
kg DM/ha of weeds compared with 57 kg DM/ha and 89 kg DM/ha for lucerne/cocksfoot
and lucerne/brome pastures, respectively. Dead material consumed was low but similar
(P<0.879) for all treatments.
50
Table 4.9 Dry matter (kg DM/ha) consumed by ewes and lambs of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes during grazing
rotation one from 5/09 to 16/10/2012 at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
(kg DM/ha)
1192 a
1088 ab
866 b
1049
81.4
0.048
Sown grass
(kg DM/ha)
49
191
120
76.5
0.247
Weed
(kg DM/ha)
140
57
89
95
53.1
0.562
Dead
(kg DM/ha)
13
12
10
12
4.10
0.879
Total
consumed
(kg DM/ha)
1345
1207
1156
1236
73.0
0.217
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Total consumed is presented as the sum of lucerne, sown grass, weed and dead
species.
4.4.2.2 Rotation 2
In grazing Rotation 2, pre-graze lucerne content of lucerne monocultures was 83%
compared with (P<0.003) 70% for lucerne/cocksfoot and 59% for lucerne/brome pastures
(Table 4.10). Lucerne monocultures had 10% weeds which was the highest (P<0.01) of all
treatments. The dead and sown grass components were not different (P<0.234) across
treatments.
Post-grazing the absolute lucerne component of monocultures declined by 0% compared
with a 21% and 16% decrease for lucerne/brome and lucerne/cocksfoot mixes,
respectively (Table 4.11). Lucerne monocultures and lucerne/cocksfoot mixes also
finished with a higher (P<0.024) lucerne content than lucerne/brome pastures. Weed
content was about 8.0% in all treatments post-grazing but sown grass content increased
to 33.5% in lucerne/brome pastures but remained at 15% in lucerne/cocksfoot pastures.
The dead material post grazing averaged 26.5% across all treatments.
51
Table 4.10 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 2 from 24/10 to
23/11/12 prior to grazing with ewes and lambs at Ashley Dene, Canterbury,
New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
82.5 a
70.1 b
58.7c
70.7
3.56
0.003
Sown grass (%)
15.8
28.2
21.9
3.88
0.07
Weed (%)
10.3 a
5.24 b
7.11 b
7.54
0.94
0.011
Dead (%)
7.25
8.14
6.00
7.13
0.83
0.234
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
Table 4.11 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for Rotation 2 from 24/10 to
23/11/12 after grazing with ewes and lambs at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
55.6 a
54.2 a
37.6 b
49.1
4.26
0.024
Sown grass (%)
15.6
33.5
24.6
4.57
0.04
Weed (%)
12.3
3.60
8.10
8.00
2.27
0.063
Dead (%)
32.1
26.6
20.8
26.5
3.69
0.146
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
There was an indication (P<0.053) that the proportion of lucerne consumed by ewes and
lambs on pure lucerne was higher than the grasses (Table 4.12). The difference was made
up of 351 kg DM/ha of brome and 219 kg DM/ha of cocksfoot in the grass mixes. Stock
grazed about 100 kg DM/ha of weeds but the dead proportion increased after grazing, by
between 120 and 220 kg DM/ha across pastures.
52
Table 4.12 Dry matter (kg DM/ha) consumed by ewes and lambs of lucerne
monocultures, lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) pastures
during Rotation 2 from 24/10 to 23/11/2012 at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
(kg DM/ha)
1618
1293
1110
1340
128.7
0.053
Sown grass
(kg DM/ha)
219
351
285
98.4
0.385
Weed
(kg DM/ha)
132
97
97
108
26.0
0.558
Dead
(kg DM/ha)
-227
-124
-119
-157
53.8
0.322
Total
consumed
(kg DM/ha)
1750
1609
1558
1639
65.7
0.140
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Total consumed is presented as the sum of lucerne, sown grass, weed and dead
species.
4.4.2.3 Rotation 3
In Rotation 3, lucerne monocultures had 73% lucerne which was higher (P<0.001) than
the 43% in the lucerne/grass mixes (Table 4.13). Weed (5%) and dead (2%) component
were low (P<0.811) for all pastures.
Post-grazing botanical composition showed lucerne content decreased by 15% to 58%
which was more than the 25% in grass mixed pastures (Table 4.14). The sown grass and
weed components remained stable (P<0.86) across treatments. The dead material
component post-grazing increased to 38% for all three treatments with weaned lambs.
Table 4.13 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for grazing Rotation 3 from
28/11/12 to 4/01/13 prior to grazing with weaned lambs at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
72.6 a
44.2 b
41.3 b
52.7
3.09
<0.001
Sown grass (%)
30.8
32.9
31.9
4.39
0.755
Weed (%)
8.03
4.47
3.77
5.40
1.39
0.116
Dead (%)
19.4
20.5
22.1
20.6
2.98
0.811
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
53
Table 4.14 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) pastures for grazing Rotation 3 from
28/11/12 to 4/01/13 after grazing with weaned lambs at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
58.1 a
24.9 b
24.8 b
35.9
2.84
<0.001
Sown grass (%)
33.2
34.3
3.37
4.08
0.86
Weed (%)
5.24
3.71
1.32
3.42
1.19
0.11
Dead (%)
36.7
38.2
39.6
38.2
4.97
0.918
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
Consumed dry matter showed the weaned lambs grazing lucerne monocultures
consumed 1207 kg DM/ha of lucerne, which was higher (P<0.017) than the 850 kg DM/ha
of lucerne consumed by weaned lambs on lucerne/cocksfoot and lucerne/brome mixes
(Table 4.15). A further 375 kg DM/ha of grass was consumed from the lucerne/grass
mixes, along with 100 kg DM/ha of weeds. Dead material increased by approximately 50
kg DM/ha across all pastures.
Table 4.15 Dry matter (kg DM/ha) consumed by weaned lambs of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes during grazing
Rotation 3 from 28/11/2012 to 4/01/2013 at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
(kg DM/ha)
1207 a
875 b
839 b
974
80.8
0.017
Sown grass
(kg DM/ha)
358
391
375
111.3
0.842
Weed
(kg DM/ha)
151
67
91
103
33.6
0.237
Dead
(kg DM/ha)
-54
-57
-39
-50
61.6
0.977
Total
consumed
(kg DM/ha)
1358
1300
1321
1326
53.3
0.712
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Total consumed is presented as the sum of lucerne, sown grass, weed and dead
species.
54
4.4.2.4 Rotation 5
In grazing Rotation 5 the lucerne monocultures contained 80% lucerne and 16% weeds
and both components were higher (P<0.018) than in the lucerne/cocksfoot and
lucerne/brome mixes (Table 4.16). In contrast, the sown grass component of
lucerne/grass mixes was ∼50%.
Table 4.16 Pre-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 5 from 15/5 to
18/06/13 prior to grazing with ewe hoggets at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
80.2 a
44.0 b
43.3 b
55.8
5.56
0.001
Sown grass (%)
50.4
48.4
49.4
1.28
0.303
Weed (%)
16.6 a
0.70 b
2.60 b
6.60
3.49
0.018
Dead (%)
3.27
4.92
5.66
4.62
0.92
0.221
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
The lucerne content decreased to 68% in lucerne monocultures and 33% in lucerne/grass
mixes (Table 4.17). However, the sown grass component increased to 54% for
lucerne/grass mixes. Weed content was highest in lucerne monocultures (P<0.009) at
25% but the dead content was lowest (7.5%).
55
Table 4.17 Post-graze botanical composition of lucerne monocultures, lucerne/cocksfoot
(Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 5 from 15/5 to
18/06/13 after grazing with ewe hoggets at Ashley Dene, Canterbury, New
Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne (%)
68.0 a
33.2 b
32.4 b
44.5
4.13
<0.001
Sown grass (%)
56.5
51.2
53.9
3.74
0.362
Weed (%)
24.6 a
0.69 b
3.69 b
9.60
4.66
0.009
Dead (%)
7.48 b
9.62 ab
12.7 a
9.94
1.03
0.015
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
Based on the yield and composition data, ewe hoggets on lucerne monocultures were
calculated to have consumed 696 kg DM/ha (P<0.011) of lucerne, compared with 378 kg
DM/ha for lucerne/cocksfoot and 396 kg DM/ha for lucerne/brome pastures (Table 4.18).
The difference in total consumption was compensated for by 375 kg DM/ha of grass
consumed in the mixes. Ewe hoggets also consumed 128 kg DM/ha of weeds in the
monoculture (P<0.047) compared with <20 kg DM/ha in the lucerne/grass mixes.
Table 4.18 Dry matter (kg DM/ha) consumed by ewe hoggets of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) or lucerne/brome (Luc/Br) mixes during Rotation 5
from 15/5 to 18/06/2013 at Ashley Dene, Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
(kg DM/ha)
696 a
378 b
396 b
490
65.6
0.011
Sown grass
(kg DM/ha)
372
378
375
12.4
0.753
Weed
(kg DM/ha)
128 a
18 b
5b
50
32.9
0.047
Dead
(kg DM/ha)
7
13
19
13
11.5
0.746
Total
consumed
(kg DM/ha)
831
781
798
803
24.0
0.469
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Total consumed is presented as the sum of lucerne, sown grass, weed and dead
species.
56
4.5 Nutritive yield
4.5.1 Metabolisable energy yield
4.5.1.1 Annual metabolisable energy yield
Annual metabolisable energy (ME) yield from sown species prior to grazing was 89 GJ
ME/ha (P<0.758) for all treatments (Table 4.19). The lucerne component of lucerne mixes
was 66 and 59 GJ ME/ha for lucerne/cocksfoot and lucerne/brome, respectively. Sown
grass of lucerne/grass mixes contributed 26 GJ ME/ha.
Table 4.19 Annual metabolisable energy (MJ ME/ha) and corresponding yield (GJ ME/ha)
and corresponding ME values (MJ/ha) of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes at Ashley Dene,
Canterbury. Sown species yield is presented as the sum of the lucerne and
sown grass yields.
Lucerne
ME
Treatment
11.0
Lucerne
10.9
Luc/CF
10.9
Luc/Br
10.9
Mean
0.0503
SEM
0.166
P value
Lucerne ME
yield (GJ/ha)
88.3 a
65.8 b
58.8 b
71.0
2.23
<0.001
Sown
grass ME
10.9
10.7
10.8
0.0573
0.073
Sown grass ME
yield (GJ/ha)
23.6
29.0
26.3
2.07
0.121
Sown species
ME yield (GJ/ha)
88.3
89.4
87.8
88.5
1.42
0.758
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
4.5.1.2 Grazing Rotation 1 metabolisable energy yield
The ME yield in Rotation 1 averaged 27 GJ ME/ha for lucerne and sown grass across all
treatments (Table 4.20).
57
Table 4.20 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) and corresponding ME values (MJ/ha) of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes for Rotation 1
from 5/09 to 24/10/12 prior to grazing with ewes and lambs at Ashley Dene,
Canterbury, New Zealand. Sown species yield is presented as the sum of
lucerne and sown grass yields.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
ME
11.6
11.6
11.7
11.6
0.0655
0.424
Lucerne ME
yield (GJ/ha)
28.2 a
24.9 ab
22.1 b
25.1
1.101
0.009
Sown
grass ME
11.6
11.4
11.5
0.1293
0.504
Sown grass ME
yield (GJ/ha)
2.74
4.08
3.41
0.670
0.218
Sown species
ME yield (GJ/ha)
28.2
27.6
26.2
27.3
1.03
0.378
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
4.5.1.3 Grazing Rotation 2 metabolisable energy yield
In Rotation 2 the ME yield of sown species averaged 26.5 GJ ME/ha (P<0.651) across all
treatments (Table 4.21). However, lucerne monocultures produced it all from lucerne
compared with 6.7 GJ ME/ha from grass in the mixes.
Table 4.21 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) and corresponding ME values (MJ/ha) of lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes for grazing
Rotation 2 from 24/10 to 28/11/12 prior to grazing with ewes and lambs at
Ashley Dene, Canterbury, New Zealand. Sown species yield is presented as the
sum of lucerne and sown grass yields.
Lucerne
Treatment
ME
11.1
Lucerne
10.7
Luc/CF
10.8
Luc/Br
10.9
Mean
0.1317
SEM
0.153
P value
Lucerne ME
yield (GJ/ha)
26.6 a
21.9 b
17.8 b
22.1
1.225
0.002
Sown
grass ME
10.7
10.4
10.6
0.100
0.108
Sown grass ME
yield (GJ/ha)
4.95
8.40
6.68
1.146
0.085
Sown species
ME yield (GJ/ha)
26.6
26.8
26.2
26.5
0.458
0.651
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
58
4.5.1.4 Grazing Rotation 3 metabolisable energy yield
In Rotation 3 the pre-grazing ME yield of all sown species in pastures was 23 GJ ME/ha
(P<0.792) (Table 4.22). In grass mixes the lucerne yield gave about 55% of the ME and the
grass 45%.
The post-graze ME yield from residuals of lucerne was 7.85 GJ/ha which was higher
(P<0.001) than the 3 GJ ME/ha from lucerne/grass mixes (Table 4.23).
Table 4.22 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) and of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for Rotation 3 from 28/11/12 to 4/01/13 prior to
grazing with weaned lambs at Ashley Dene, Canterbury, New Zealand.
Lucerne ME
yield (GJ/ha)
22.8 a
13.0 b
13.1 b
16.3
0.811
<0.001
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Sown grass ME
yield (GJ/ha)
9.19
10.2
9.70
1.321
0.604
Sown species
ME yield (GJ/ha)
22.8
22.3
23.3
22.8
1.13
0.792
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
Table 4.23 Post-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) of lucerne, lucerne/cocksfoot and lucerne/brome pastures for grazing
Rotation 3 from at Ashley Dene, Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
ME
7.3
6.8
7.1
7.1
0.200
0.241
Lucerne ME
yield (GJ/ha)
7.85 a
2.91 b
3.53 b
4.76
0.497
<0.001
Sown
grass ME
10.3
9.2
9.7
0.218
0.015
Sown grass ME
Sown species
yield (GJ/ha) ME yield (GJ/ha)
7.85
5.73
8.64
6.06
9.59
5.90
8.69
0.686
1.049
0.745
0.295
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
59
There was no difference (P<0.521) in the total (14.1 GJ/ha) of ME consumed from sown
species by weaned lambs across all treatments (Table 4.24). Weaned lambs on grass
mixes consumed 3.82 GJ ME/ha from grass and the rest from lucerne.
Table 4.24 Metabolisable energy consumed by weaned lambs grazing lucerne,
lucerne/cocksfoot and lucerne/brome pastures in early summer at Ashley
Dene, Canterbury, New Zealand.
Lucerne
(GJ ME/ha)
14.9 a
10.1 b
9.57 b
11.5
0.735
<0.001
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Sown grass
(GJ ME/ha)
3.47
4.17
3.82
1.516
0.664
Sown species
(GJ ME/ha)
14.9
13.6
13.7
14.1
0.861
0.521
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests.
Sown species consumed is presented as the sum of the lucerne and sown grass yields.
4.5.1.5 Grazing Rotation 5 metabolisable energy yield
In autumn, sown species ME yields were highest (P<0.049) from lucerne/cocksfoot
predominantly due to the 6.67 GJ ME/ha from grass (Table 4.25).
Table 4.25 Pre-grazing metabolisable energy (MJ ME/ha) and corresponding yield (GJ
ME/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 5 at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
ME
11.4
11.2
11.2
11.3
0.0616
0.159
Lucerne ME
yield (GJ/ha)
10.7 a
5.91 b
5.80 b
7.47
0.701
0.001
Sown
grass ME
11.3
11.5
11.4
0.0889
0.192
Sown grass ME
yield (GJ/ha)
6.67
6.31
6.49
0.219
0.299
Sown species
ME yield (GJ/ha)
10.7 b
12.6 a
12.1ab
11.80
0.482
0.049
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
60
4.5.2 Nitrogen (N) yield
4.5.2.1 Annual nitrogen yield
The total annual N yield from lucerne was 286 kg/ha from lucerne monocultures (Table
4.26). There was an indication (P<0.073) that this was higher than from lucerne/brome.
Table 4.26 Annual nitrogen concentration (N%) and corresponding annual nitrogen yield
(GJ ME/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) pastures at Ashley Dene, Canterbury. Sown species
yield is presented as the sum of lucerne and sown grass yields.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
N%
3.7
3.6
3.6
3.6
0.0534
0.605
Lucerne N
yield (kg/ha)
286a
213 b
189 b
229
10.83
<0.001
Sown
grass N%
3.2
2.8
3.0
0.1163
0.070
Sown grass N
yield (kg/ha)
63.8
68.4
66.1
5.45
0.569
Sown species
N yield (kg/ha)
286
277
257
273
7.78
0.073
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
4.5.2.2 Grazing Rotation 1 nitrogen yield
In Rotation 1 there was no difference (P<0.689) in N yield (87.1 kg N/ha) from lucerne
and sown grass components across all treatments (Table 4.27).
Table 4.27 Pre-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 1 at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
N%
3.7
3.8
4.0
3.8
0.1723
0.509
Lucerne N
yield (kg/ha)
89.7
80.1
74.8
81.5
4.86
0.138
Sown
grass N%
3.0
2.8
2.9
0.0733
0.169
Sown grass N
yield (kg/ha)
6.83
9.98
13.7
1.61
0.225
Sown species
N yield (kg/ha)
89.7
87.0
84.8
87.1
4.01
0.689
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
61
4.5.2.3 Grazing Rotation 2 nitrogen yield
In Rotation 2, the lucerne component of lucerne monocultures produced 86 kg N/ha,
which was higher (P<0.013) than the 69 kg N/ha and 54 kg N/ha produced by lucerne in
lucerne/cocksfoot and lucerne/brome mixes, respectively (Table 4.28). There was no
difference (P<0.819) in sown grass nitrogen yield between lucerne/cocksfoot and
lucerne/brome mixes. The sown species nitrogen yield was not different (P<0.114)
between treatments.
Table 4.28 Pre-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 2 at Ashley Dene,
Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
N%
3.6
3.3
3.3
3.4
1.224
0.269
Lucerne N
yield (kg/ha)
85.7 a
68.6b
54.3 b
69.5
4.61
0.003
Sown
grass N%
3.1
2.6
2.8
0.248
0.220
Sown grass N
yield (kg/ha)
13.4
18.9
16.2
1.299
0.819
Sown species
N yield (kg/ha)
85.7
82.0
73.3
80.3
3.85
0.114
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
4.5.2.4 Grazing Rotation 3 nitrogen yield
Lucerne in lucerne monocultures produced the highest (P<0.001) N yield of 69 kg/ha
compared with 39 kg/ha for lucerne/grass mixes (Table 4.29). However, there was no
difference (P<0.462) in the N yield of sown grass in the lucerne/grass mixes, or the total N
across treatments.
There was also no difference (P<0.749) in sown species nitrogen yield across all
treatments post-grazing (Table 4.30). The lucerne component of lucerne monocultures
yielded 17 kg N/ha after grazing which was higher (P<0.005) than the ∼ 7 kg N/ha
yielded by both the lucerne/grass mixes.
62
Table 4.29 Pre-grazing nitrogen concentration (N%) and nitrogen yield (kg/ha) of lucerne
monocultures, lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes
for Rotation 3 from 28/11/12 to 4/01/13 prior to grazing with weaned lambs at
Ashley Dene, Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
N%
3.0
3.1
2.8
3.0
0.1479
0.310
Lucerne N
yield (kg/ha)
68.5 a
41.3 b
37.1 b
49.0
4.16
<0.001
Sown
grass N%
2.5
1.8
2.2
0.1306
0.016
Sown grass N
yield (kg/ha)
21.5
19.5
20.5
2.52
0.462
Sown species
N yield (kg/ha)
68.5
62.8
56.6
62.6
4.47
0.222
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
Table 4.30 Post-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 3 from 28/11/12 to 4/01/13
after grazing with weaned lambs at Ashley Dene, Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
N%
1.6
1.5
1.5
1.5
0.0568
0.549
Lucerne N
yield (kg/ha)
17.1 a
6.71 b
7.54 b
10.45
1.021
<0.001
Sown
grass N%
2.1
1.6
1.8
0.0751
0.126
Sown grass N
yield (kg/ha)
11.7
10.8
11.25
1.004
0.546
Sown species
N yield (kg/ha)
17.1
18.4
18.3
17.9
1.324
0.749
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
As a consequence, there was no difference (P<0.163) in the nitrogen consumed from
sown species across all treatments (Table 4.31). Weaned lambs grazing lucerne
monocultures consumed 51 kg N/ha from lucerne which was greater (P<0.005) than from
lucerne in lucerne /grass mixes. An additional 9.28 kg N/ha was consumed from both
(P<0.836) sown grasses.
63
Table 4.31 Nitrogen consumed (kg N/ha) by weaned lambs grazing lucerne monocultures,
lucerne/cocksfoot (Luc/CF) and lucerne/brome (Luc/Br) mixes during Rotation
3 from 28/11/12 to 4/01/13 at Ashley Dene, Canterbury, New Zealand.
Lucerne
(kg N/ha)
51.4 a
34.6 b
29.6 b
38.5
3.78
0.005
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Sown grass
(kg N/ha)
9.79
8.77
9.28
3.32
0.836
Sown species
(kg N/ha)
51.4
44.3
38.3
44.7
4.42
0.163
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species consumed is presented as the sum of the lucerne and sown grass
yields.
4.5.2.5
Grazing Rotation 5 nitrogen yield
The nitrogen yield consumed in rotation five showed no differences (P<0.416) across all
treatments (Table 4.32).
Table 4.32 Pre-grazing nitrogen concentration (N%) and corresponding nitrogen yield
(kg/ha) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes for grazing Rotation 5 from 15/5 to 18/06/13
prior to grazing with ewe hoggets at Ashley Dene, Canterbury, New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Lucerne
N%
4.5
4.4
4.4
4.4
0.0623
0.471
Lucerne N
yield (kg/ha)
42.0
23.3
22.7
29.3
2.73
<0.001
Sown
grass N%
4.1
4.0
4.0
0.1375
0.552
Sown grass N
yield (kg/ha)
22.1
20.0
21.1
0.913
0.987
Sown species
N yield (kg/ha)
42.0
45.4
42.7
43.4
1.85
0.416
Note: Treatment means followed by the same letter are not significantly different at α=0.05 using least
significant difference tests. Sown species yield is presented as the sum of the lucerne and sown grass yields.
64
4.6 Soil water content
4.6.1 Available water
The available soil water for each plot was calculated by determining the drained upper
limit (DUL) and lower limit (LL) (Section 3.7). The total water available over the growing
season was determined by calculating the sum of the available water in each layer of the
soil profile down to 225 mm. The highest value of 248 mm of available water was in plot
7 and the lowest of 123 mm in plot 2 (Figure 4.7). These values represent the highest and
lowest total plant available water across all 18 plots. For other plots refer to Appendix 1.
The amount of soil available varied down the soil profile, due to the soil texture. In the
top 0.2 m plot 7 had 77 mm of water available to lucerne and brome plants while plot 2
had 90 mm of water available to lucerne and cocksfoot plants. For plot 7, from 0.2 m to
2.05 m the profile plant available water ranged from 26.6 mm/0.2 m soil to 13.1 mm/0.2
m of soil compared with 21.6 mm/0.2 m of soil to 2.6 mm/0.2 m of soil for plot 2. The
plant available water in plot 2 from 0.65 m to 2.25 m in plot 2, was <10 mm/0.2 m of soil.
At the lowest recorded depth of 2.3 m, the plant available water was 7.1 mm/0.2 m of
soil for plot 7 compared with 1.9 mm/0.2 m of soil for plot 2.
PAWC was not different between treatments (P>0.05), with a mean of 183 mm. This
explains why dry matter yield was not different (P<0.001) due to the same amount of
water being available. However, the PAWC between plots was highly variable ranging
from 123 mm in plot 2 to 248 mm from plot 7. This is the result of differences in soil type
and texture. Plot 7 had more plant available water due to there being more sand and silt
present in the lower part of the soil profile compared with plot 2 which had lots of
stones, therefore little plant available water. However, plot 2 had greater plant available
water in the top 0.5 m of the soil profile. This could be due to organic matter and silt in
the topsoil.
65
Figure 4.7 Water extraction pattern of lucerne, cocksfoot and brome roots in the soil
profile. Where (●) is the upper limit and (○) is the lower limit for plant
available water in Plot 7 (top) and Plot 2 (bottom) in paddocks C6E and C7W,
Ashley Dene, Canterbury.
66
4.6.2 Soil water content
Rainfall and soil water content (SWC) over time are shown in Figure 4.8. Measurements
began on the 26/11/12. On this date plot 7 had a soil water content of 371 mm compared
with 262 mm for plot 2. The soil water content declined from then on, due to
evapotranspiration exceeding rainfall. The lowest soil water content for both plots was
reached on the 13/03/2013 with 209 mm and 181 mm of soil water for plots 7 and 2,
respectively. After this point rainfall exceeded evapotranspiration and soil water content
increased. During May and June 2013 there was a total of 316 mm of rainfall, increasing
the soil water content. The maximum soil water content for plot 2 of 296 mm was
reached on 28/05/13. Plot 7 reached its maximum soil water content of 469 mm on the
30/06/13. On this date the soil water content of plot 2 had decreased to 253 mm.
Despite the plot to plot variability, there was no difference (P<0.263) in the amount of
water used across treatments for the year 1 July 2012 to 30 June 2013 (Table 4.33). All
pastures used 612 mm of water.
Table 4.33 Water use (WU) of lucerne monocultures, lucerne/cocksfoot (Luc/CF) and
lucerne/brome (Luc/Br) mixes from 1/07/12 to 30/06/13 at Ashley Dene, Canterbury,
New Zealand.
Treatment
Lucerne
Luc/CF
Luc/Br
Mean
SEM
P value
Water use (mm)
569
684
584
612
50.5
0.263
67
100
400
80
300
60
200
40
100
20
0
Jul12
Oct12
Jan13
Apr13
Rainfall (mm)
Soil water content (mm)
500
0
Jul13
Time
Figure 4.8 Soil water content (mm) and rainfall (mm) for Plot 7 (—) and Plot 2 (—) in
paddocks C6E and C7W, Ashley Dene, Canterbury. Rainfall data are taken from
the Ashley Dene weather station (43°65’S, 172°32’E).
4.6.3 Water use efficiency
The WUE of lucerne monocultures, lucerne/cocksfoot and lucerne/brome mixes was
calculated by plotting the cumulative dry matter against cumulative water use and fitting
a linear regression (Figure 4.9). The regression was fitted to the period when ewes and
lambs grazed as water was non-limiting during this period. There was no difference
(P>0.05) in WUE among treatments therefore a single regression was fitted to the data.
All treatments had a spring WUE of 22±0.11 kg DM/ha/mm of water.
68
Accumulated dry matter yield (kg DM/ha)
12000
10000
8000
6000
4000
2000
0
0
100
200
300
400
500
600
700
Accumulated water use (mm)
Figure 4.9 Relationship between accumulated dry matter yield (kg DM/ha) and
accumulated water use (mm) for lucerne monocultures (●), lucerne/cocksfoot
(■) and lucerne/brome (▽) mixes. Form of the spring regression line is:
22.0±0.11x + 85.5±29.6 (R2=0.99).
69
5
DISCUSSION
5.1 Animal production
5.1.1 Annual and seasonal liveweight production
Annual liveweight production was 16% greater from lucerne monocultures and
lucerne/cocksfoot compared with lucerne/brome (Figure 4.1). Lucerne monocultures and
lucerne/cocksfoot mixes produced 865 kg LWT/ha compared with lucerne/brome mixes
which produced 746 kg LWT/ha. This is comparable with results from Brown et al. (2006)
who showed with liveweight production over two years of 880 kg/ha for dryland lucerne
monocultures.
The highest liveweight production was during spring with approximately 70% of the
annual liveweight production occurring for all three treatments. Spring dry matter
production also accounted for ∼50% of the accumulated total. These results were
expected as soil moisture is not limiting pasture growth during this time (Brown et al.,
2006). Mills et al. (2008b) determined that over five years, an average of 64% of the
annual liveweight production occurred during spring. Spring livestock production
(September-November) for lucerne monocultures was 570 kg LWT/ha, 584 kg LWT/ha for
lucerne/cocksfoot and 537 kg LWT/ha for lucerne/brome mixes. This was well above the
400 kg LWT/ha recorded by Brown et al. (2006) for dryland lucerne monocultures during
spring (July-November). This experiment produced more liveweight gain in a shorter
period than Brown et al. (2006), therefore differences were probably due to stocking
rates. They used a group of ‘core’ hoggets and added more to match feed supply with
demand, however stocking rates were not specified. A higher stocking rate would
increase the amount of liveweight production. Summer (December-January) production
of 252 kg LWT/ha for lucerne monocultures and 236 kg LWT/ha for lucerne/cocksfoot
was higher than the 149 kg LWT/ha for lucerne/brome mixes. This was less than half the
production of 550 kg LWT/ha for lucerne monocultures recorded by Brown et al. (2006)
for summer (December-January). This large difference in liveweight production between
the two experiments is either due to quantity or quality of pasture produced. Pasture
quality was the same in both experiments, but dry matter production was different.
70
Brown et al. (2006) had dry matter yields >16t DM/ha compared with only 12 t DM/ha for
the current experiment (Figure 4.4). This resulted in a longer summer grazing duration for
Brown et al. (2006) of three months compared with two months (Section 3.8) at Ashley
Dene.
5.1.2 Rotational liveweight gains
The difference in liveweight gain can only be a result of differences in per head
performance or stocking rate. During liveweight Rotations 1, 2 and 3, lambs pre-weaning
maintained average daily liveweight gains of 300 g/head/d across all treatments (Table
4.2). These results are comparable with Douglas et al. (1995) who recorded average
liveweight gains of 263 g/head/d for unweaned lambs grazing lucerne. However, daily
liveweight gains are only comparable if stocking rates are the same, otherwise
differences in feed allocation arise. Stocking rates were not specified by Douglas (1995),
therefore it is not known if these comparisons are realistic, as liveweight gains in their
experiment could be a result of greater pasture allocation.
During the second liveweight rotation, ewes had an average daily growth rate of 212
g/head/d on lucerne monocultures which was higher than the 170 and 181 g/head/d on
lucerne/cocksfoot and lucerne/brome (Table 4.2). Thus, lucerne monocultures produced
superior liveweight gain probably due to higher herbage quality (Section 5.4). The lower
liveweight gains of ewes in the third liveweight rotation could have been due to
increasing lamb demand. By the third rotation, lambs would be more demanding of milk
and pasture. Ewes in liveweight rotation three were calculated to have required 44 MJ
ME/day to sustain them and their lambs which was 47% more than the 33 MJ ME/day
required by ewes and their lambs during rotation one (Appendix 1). This indicates that
the ewes were sacrificing their liveweight gains to maintain lamb liveweight gains of
∼300 g/head/d throughout grazing Rotations 1 and 2 (Figure 4.2).
Liveweight Rotation 4 was where the main advantage of lucerne was expressed. Weaned
lambs for this experiment gained 243 g/head/d on lucerne monocultures, compared with
221 g/head/d on lucerne/cocksfoot and 164 g/head/d on lucerne/brome mixes. This is
more than the liveweight gains of 182 g/head/day for weaned lambs recorded by Douglas
71
et al. (1995) and 160 g/head/d recorded by Mills et al. (2008b). However, caution is
needed in comparing experiments because stocking rates were not specified therefore
differences in daily liveweight gains could be explained by different stocking rates rather
than herbage quality or quantity.
Ewe hoggets during liveweight Rotation 5 had the highest gains on lucerne/brome
pastures at 109 g/head/day. Stocking rates on lucerne/brome pastures were also higher
than other treatments. Therefore, more feed was available for livestock in the
lucerne/brome than other treatments at this time (Section 3.8).
5.2 Pasture dry matter yield
5.2.1 Accumulated dry matter yield
The ability to maintain a higher stocking rate on lucerne monocultures and
lucerne/cocksfoot was either because of greater dry matter production or higher herbage
quality. Accumulated dry matter yields were not different and ranged from 12.8 t/ha for
lucerne monocultures to 12.2 t/ha for lucerne/grass (Figure 4.4). Therefore, differences in
animal liveweight gain were not due to the amount of pasture available. Accumulated dry
matter yields in this experiment was higher than yields recorded at Invermay Research
Station (Cullen, 1965). Their lucerne monocultures produced 6.5 t DM/ha compared with
9.6 and 12 t DM/ha for lucerne/cocksfoot and lucerne/brome pastures, respectively. This
different could be due to climatic conditions (Section 2.5). However, current yields were
lower than those recorded elsewhere in Canterbury. Lucerne monocultures sown in
Lincoln produced 19.5 t DM/ha compared with 19 t DM/ha for lucerne/brome pastures
(Fraser, 1982) . Sowing date and rate were similar between both trials with 7-8 kg/ha of
lucerne and 10 kg/ha of brome sown in September or October. Differences in dry matter
yield are probably the result of differences in water use and how efficiently the water
extracted was used. Soil water holding capacity determines the amount of plant available
water and varies according to soil type.
72
5.2.2 Mean daily growth rates
All pastures grew at 30 kg DM/ha/d from 1/07 to 5/09/12 and then increased to a
maximum in November. During this time lucerne/brome grew at the highest (P<0.007)
rate of 98 kg DM/ha/d and lucerne/cocksfoot the lowest at 89 kg DM/ha/d (Section
4.2.2). After this evapotranspiration exceeded rainfall and growth rates declined to 22 kg
DM/ha/d (Section 0). Further declines to 16 kg DM/ha/d were calculated for March to
May 2013. These results are comparable with studies carried out at Lincoln University,
Canterbury, New Zealand. Brown (2003) recorded mean daily growth rates for lucerne
monocultures to be 34 kg DM/ha/d during spring which increased to a maximum of 90 kg
DM/ha/d during December and January. Growth rates declined from then on to 20 kg
DM/ha/d in May. Tonmukayakul et al. (2009) also recorded a maximum growth rate of
100 kg DM/ha/d for lucerne monocultures in November which declined to <10 kg
DM/ha/d in June. A maximum mean daily growth rate of 92 kg DM/ha/d was reported for
lucerne monocultures in Canterbury during December which was very similar to this
experiment (Mills and Moot, 2010). Having no difference in yield and mean daily growth
rates indicates that differences in liveweight gains were not a result of more pasture
supply. Therefore they could be due to higher herbage quality.
5.3 Soil water budget
Differences in dry matter production are either due to more plant available water, or
greater water use efficiency. The mean water used (WU) for all treatments was 612 mm.
The mean water use efficiency (WUE) was also the same for all treatments with 22 kg
DM/ha/mm. There could have been a difference in both WU and WUE however it could
have been masked due to the large variability between individual plots. No differences in
WU and WUE explain why there were no yield differences. The same amount of water
was used at the same efficiency for all treatments. The water used was nearly double the
384 mm reported by McKenzie et al. (1990) for lucerne/brome mixes. However, the
water use for the experiment ranged from 396 mm to 812 mm for individual plots. The
dry matter yields of pastures are low in comparison with other literature (Brown et al.,
2006; Tonmukayakul et al., 2009). This is due to the two different soil types having
different plant available water values.
73
5.4 Thermal time
The rate of pasture supply was also constant across treatments when calculated on a
thermal time basis. All treatments grew at an average rate of 5.5 kg DM/ha/°Cd during
spring (Figure 4.6). After this, rates declined to 1.07 kg DM/ha/°Cd for lucerne/cocksfoot
compared with 0.82 and 0.68 kg DM/ha/°Cd for lucerne monocultures and
lucerne/brome, respectively. This shows there were no differences in accumulated dry
matter yield due to their being a temperature response. During the second phase, from 5
March and 15 May 2013, the pastures were severely water stressed (Section 4.6.2). The
spring growth rates are slightly higher than the 4.9 kg DM/ha/°Cd reported by
Tonmukayakul et al. (2009) for lucerne monocultures. Autumn growth rates from
Tonmukayakul et al. (2009) for lucerne monocultures were 1.1 kg DM/ha/°Cd which was
higher than calculated in this experiment. Therefore, temperature had a greater effect on
dry matter production in Tonmukayakul et al. (2009). This was due to the soil type of the
experiment, resulting in more plant available water resulting in greater dry matter
production (Section 5.2.1).
5.5 Botanical composition
Annually, lucerne monocultures had at least 20% more lucerne and 5% more weeds than
lucerne/grass mixes pre-grazing. Post-grazing, the lucerne in pastures had declined
relatively more than other components by ∼12% for all treatments (Table 4.5,Table 4.6).
In contrast, the dead content of all pastures doubled after grazing while the proportion of
sown grass content remained the same. This suggests sheep were grazing selectively for
lucerne, and against sown grasses. It is known that sheep exhibit preferential grazing for
forage high in nitrogen (Edwards et al., 1993) therefore lucerne was probably selected for
rather than nitrogen deficient sown grasses and dead material. The lucerne component
of the lucerne/grass mixes was nearly double that of Cullen (1965) who recorded lucerne
contents of 53% in lucerne monocultures, 20% in lucerne/cocksfoot and 7% in
lucerne/brome. This could be explained by grazing management, with infrequent grazing
periods favouring lucerne growth, while frequent grazing favouring grass growth (Section
2.7).
74
In Rotation 2, lucerne content of all pastures decreased as a result of ewes and lambs
consuming more lucerne than any other component (Table 4.9). Sown grass content
increased in lucerne/brome but remained the same in lucerne/cocksfoot. Dead material
increased by an average of 20% for all treatments. This also suggests that grazing
selection for lucerne was occurring and against brome grass and dead material. The
implication is that the consumed diet was of higher quality than what was on offer (Table
4.21). Post-graze nutritive values would have determined exactly the quality of the diet
that stock selected. Stock on lucerne monocultures also consumed 100 kg DM/ha of
weeds, which could have been to obtain sodium, due to lucerne having a low sodium
content (Joyce and Brunswick, 1975).
Lucerne monocultures in Rotation 3 also had a higher lucerne content than lucerne/grass
both pre and post-grazing (Table 4.13,Table 4.14). Lucerne content declined by 15% in
lucerne monocultures and 18% in lucerne/grass mixes. Dead material increased by 50
kg/ha for all treatments, again emphasizing the effect of diet selection. The increase in
dead material could have been caused by trampling, senescence or the presence of seed
heads from reproductive grass plants.
The lucerne content of all pastures declined by ∼12% after autumn grazing with ewe
hoggets (Table 4.17). The sown grass content increased to 54% in lucerne/grass mixes,
making them grass dominant for the first time during the experimental period. This was
because lucerne had become dormant over winter while brome remained active and
cocksfoot also showed some growth. This difference in the temporal pattern of pasture
growth is one reason why lucerne/grass mixes are looked at as an alternative to pure
lucerne stands. Ewe hoggets on lucerne monocultures also consumed 100 kg DM/ha
more of weeds than those on lucerne/grass (Table 4.18), highlighting a possible desire for
a varied diet to balance mineral intake (Joyce and Brunswick, 1975).
5.6 Nutritive yield
5.6.1 Metabolisable energy
All treatments produced 88 GJ ME/ha from sown species (Table 4.19). Lucerne
monocultures produced all of this from lucerne while sown grasses contributed 26 GJ
75
ME/ha to the annual ME production in the grass mixes. This indicates that differences in
annual liveweight production were not a result of the quality of pastures. These annual
yields are lower than the ∼134 GJ ME/ha yield reported by Mills and Moot (2010) for
lucerne monocultures grown at Lincoln University, Canterbury, New Zealand. Their dry
matter yields were 14.0 t DM/ha which was also higher than those reported for the
experiment (Section 4.2.1). This additional yield is the main reason for differences in ME
yield, as the herbage quality for both experiments was the same.
During all grazing rotations there was no difference in the herbage quality of sown
species across all treatments (Section 4.5.1). There was also no difference in herbage
quality when compared with other literature (Brown et al., 2006). Lucerne monocultures
produced more ME yield from the lucerne component than lucerne/grasses but the
contribution of grass to the ME yield, compensated for this. Differences in sown ME yield
across rotations reflected the different growth patterns of the lucerne and grasses. The
highest ME yields of 27 GJ ME/ha were recorded for grazing rotation one, when water
was not limiting plant growth (Table 4.20).
During grazing rotation three, weaned lambs consumed 14 GJ ME/ha for all treatments
(Table 4.24). However, there was a difference in liveweight gains during this rotation. This
indicates that differences were not due to the quality of feed intake but rather due to
differences in grazing days (Section 5.6.3). Post-graze nutritive values give an exact
indication of what livestock consumed, as grazing selection means the quality offered is
not always the same as what is consumed, with livestock favouring forage high in N
(Section 2.5).
The lowest ME yield was recorded pre-grazing in rotation five with 11.8 GJ ME/ha for all
treatments (Table 4.25). This could be explained by the low winter activity of lucerne with
sown grass species contributing 6.5 GJ ME/ha compared with 5.9 GJ ME/ha for lucerne in
the lucerne/grass mixes. Therefore, even though quality remained constant, dry matter
yield was lower during autumn resulting in lower ME yields.
76
5.6.2 Nitrogen yield
Annual nitrogen yield was also the same for all treatments, producing 286 kg/ha (Table
4.26). Like ME yield, lucerne monocultures produced more N from lucerne than
lucerne/grass treatments (286 kg/ha versus 201 kg/ha). This further indicated that
differences in liveweight production were not a result of pasture quality. These results
are lower than the 410 kg N/ha produced by lucerne monocultures at Lincoln University,
Canterbury, New Zealand (Tonmukayakul et al., 2009). However, differences can be
explained by pasture yield. Tonmukayakul et al. (2009) figures were from lucerne stands
established on a Templeton silt loam which has greater plant available water than the
Lismore, Lowcliff and Ashley Dene soils in this experiment (Section 5.2.1).
Pre-grazing with weaned lambs, there were no differences in the N yield across pastures
(69 kg/ha) (Table 4.29). However, lucerne monocultures produced more N yield than the
lucerne in lucerne/grass mixes. As with the ME yield, the sown grass N contribution
resulted in no difference in the sown component N yield. During grazing rotation three,
weaned lambs consumed 44 kg N/ha for all treatments (Table 4.31). Tonmukayakul et al.
(2009) reported an N yield of 145 kg/ha for established lucerne stands during summer.
Differences could be explained by the length of the period, grazing rotation was 44 days
in duration compared with 111 days for Tonmukayakul et al. (2009).
The lowest N yield was in grazing rotation five with 43 kg/ha for all treatments. This was
comparable with the 39 kg N/ha recorded by Tonmukayakul et al. (2009) for lucerne
monocultures during autumn/winter. This further indicated that differences in liveweight
production were not a result of quality of the pastures.
5.6.3 Production graze days
There were no differences in daily growth rates, dry matter yield and herbage quality,
therefore differences in liveweight gain must be due to production days.
Lucerne/cocksfoot and lucerne monocultures had 3855 and 3810 production graze days
for the year which was more than lucerne/brome with 3554 (Table 4.1). Lucerne/brome
mixes had comparable liveweight gains with lucerne monocultures and lucerne/cocksfoot
however the number of graze days was lower. This indicates that lucerne/brome pastures
77
were unable to sustain the same number of livestock as lucerne monocultures and
lucerne/cocksfoot pastures. Stocking rates were lower on the lucerne/brome pastures
during grazing rotations two and three (Table 3.4) than other treatments, explaining why
the number of production days was different. However, there was no difference in the
total number of grazing days due to lucerne/brome being stocked at 27.9 SU/ha during
grazing rotation four. This meant that lucerne/brome accumulated more maintenance
grazing days that the other treatments.
The number of production days for all treatments are well above the production grazing
days for lucerne monocultures as part of the ‘Maxclover’ experiment which averaged
1482 production days over five years (Mills et al., 2008a). The difference between the
two experiments resulted because previously reported data was per plot not per hectare
as specified. This explains why although DM yield was greater from ‘Maxclover’ grazing
days were lower than those reported here. Liveweight gains were also similar for both
experiments (Section 2.2), however hoggets were used for ‘Maxclover’ compared with
ewes and lambs for this experiment.
78
5.7 Conclusions
•
Liveweight
production
was
greater
from
lucerne
monocultures
and
lucerne/cocksfoot than lucerne/brome
•
Daily liveweight gains for livestock were generally similar across all pastures,
however stocking rates were different, resulting in differences in liveweight
production/ha.
•
There were no differences in dry matter yield across all pastures. This was the
result of no differences in water use, water use efficiency and thermal time.
•
Pre- and post-graze botanical composition indicated that diet selection for lucerne
and against sown grass and dead material occurred, particularly in lucerne/brome
pastures.
•
Grazing management of the lucerne/brome pasture was no effective in
maintaining pasture quality throughout the growing season.
•
There was no difference in ME and N content and yield between pastures,
therefore liveweight production differences were a result of stocking rate, not
quantity or quality of herbage.
79
6
GENERAL DISCUSSION AND CONCLUSIONS
6.1 General discussion
6.1.1 Establishment
Lucerne/grass mixes can be difficult to establish, and this was the case with this
experiment. Both grass species had to be resown due to poor emergence. Poor
establishment of herbage species can occur if they are sown below 10-15 mm. the initial
sowing method was rushed and had machinery malfunctions (M. Smith, 2013. Personal
communication) which necessitated redrilling. Care with sowing and early establishment
of lucerne grass mixes is needed to maximise subsequent pasture production.
6.1.2 Sodium content
The consequent, livestock production on these pastures indicated superior performance
from lucerne and lucerne with cocksfoot. Livestock production on lucerne monocultures
has also proven superior than other grassland pastures in dryland conditions (Brown et
al., 2006; Mills et al., 2008b; Mills and Moot, 2010). However, planting an entire farming
system in lucerne is generally not suitable due to is low winter activity, which can result in
a feed deficiency. Also, lucerne has a low sodium (Na) content, which is often below
animal requirements (Sherrell, 1984). In this experiment with livestock grazing lucerne
consumed more weeds than livestock grazing lucerne/grass mixes which may indicate
their desire to require salt (Section 4.4). However this could be overcome by
supplementation of salt blocks and it is surprising this is not being done. Sowing a
companion grass that has adequate Na levels may reduce the need for supplementation.
Despite the lack of Na supplementation the lucerne monocultures had as much animal
and plant production as the mixtures. This was because the WUE, feed quality and
temperature response of lucerne was as high as the lucerne but superior to
lucernre/brome. Thus, results from this first year of production suggest either the lucerne
monoculture or lucerne/cocksfoot should be sown in preference to lucerne/brome. The
main limitation of the lucerne/brome was its early reproductive development. This meant
it was stocked at lower rates than the other two pastures in the main spring period of
animal liveweight gain. The additional grazing days it required during summer were to
80
‘cleanup’ the excessive seed head material with maintenance stock. This feed would not
have been suitable for production stock but quality estimates were not taken. A
recommendation from this study would be to measure both the quality and quantity of
this feed eaten during mid-summer.
6.1.3 Effect of nitrogen on pastures
Despite the high production from lucerne monocultures there may be other reasons to
consider planting a lucerne/grass mix. Lucerne has a high nitrogen content (Section
4.5.2), which means urine N of livestock grazing it are likely to be high. This could lead to
high returns to the soil and nitrate leaching over time (Haynes and Williams, 1993).
Planting a companion grass species with lucerne, may allow the grass to utilise this
available N and reduce soil N levels. Cocksfoot is strongly summer active but is generally
considered unpalatable to livestock. The addition of nitrogen from these urine returns
appears to have increased both the quantity, quality and the palatability of cocksfoot.
Yields of 28 t DM/ha have been reported for cocksfoot pastures with nitrogen applied
(Mills et al., 2006) so the urine returns may have aided cocksfoot production in these
mixes. The pattern of livestock grazing observed indicated animals grazed the lucerne
upon entry to the paddock and then found the cocksfoot. This differential in diet
selection may reduce overtime as more nitrogen is available to the cocksfoot from urine
returns.
The addition of nitrogen to the pasture system is predominantly from N fixed by the
lucerne being cycled via the animal to the grass plants. Thus, N application represents a
considerable saving in production costs for dryland farmers. Nitrogen could be added via
artificial fertiliser. Urea (46,0,0,0) is currently selling for $640 t exclusive of GST (J. Coutts,
2013. Personal communication). Fasi et al. (2008) reported cocksfoot responded at a rate
of 19.4 kg DM/kg N to fertiliser applications. Based on the current cost of urea, this
equates to a cost of $0.15 per kg DM grown (($640*0.46)/19.4). This is economic for
dryland farmers to apply urea to pastures, however many dryland farmers do not use it.
Cocksfoot has successfully been sown with subterranean clover, with yields of 9.8 t
DM/ha resulting in liveweight production of 912 kg/ha (Mills and Moot, 2010). Based on
the lucerne yields from this experiment the fixed nitrogen added to each pasture could
81
be estimated as 410 kg for lucerne monocultures and 295 and 257 kg from the lucerne in
the cocksfoot and brome, respectively.
6.1.4 Grazing management
Other, legumes such as sub clover could also be sown as a method of supplying nitrogen.
Most legume species seem to fix ∼25 kg N/kg DM grown above ground annually (Lucas
et al., 2010). In that ‘Maxclover’ experiment cocksfoot was also twice as persistent as
perennial ryegrass. Thus the sowing of lucerne as the companion legume with cocksfoot
may provide productive and persistent dryland pastures. However, sowing cocksfoot with
lucerne increases the complexity of grazing management due to both species having
different best management practices. Lucerne stands are best rotationally grazed, with
infrequent grazing periods due to the elevated position of the growing point. In contrast,
cocksfoot has an aggressive growth pattern and is suited to heavy defoliation to maintain
a leafy vegetative plant (Mills et al., 2006). Lucerne/cocksfoot pastures can either be
managed as a lucerne stand with cocksfoot sown or as a cocksfoot pasture with lucerne.
Grazing management determines whether the pasture will be cocksfoot dominant or
lucerne dominant. Frequent, heavy defoliation will favour cocksfoot growth, while
infrequent defoliation will favour lucerne growth (Cullen, 1965). It may also be possible to
change the balance of the species within such a pasture by infrequent grazing in spring,
early summer and autumn, but frequent grazing in mid-summer to stop cocksfoot
becoming clumpy.
Brome had less liveweight production than the other two pastures, however this does not
mean it should be dismissed as a companion grass species for lucerne. Dead material in
those pastures also built up during summer months, due to it not being selected by ewes
and lambs. However, this could have just been due to the grazing management used
during the experiment. Lucerne/brome was stocked at a lower rate than the other
pastures which resulted in a built up of seed heads and dead material. In a commerical
farming system, a leader and follower method of grazing could be implemented. Ewes
and lambs could be used as the leaders to graze pastures first, and cattle could then
follow behind to clean up the pastures. Cattle are less selective at grazing than sheep,
due to jaw morphology (Grant et al., 1985). Therefore, changes in botanical composition
82
as a result of grazing selection could be decreased, with an exception that more of the
early reproductive brome stem would be consumed. This could aid in maintaining the
pasture quality. Equally, lamb liveweight gains would be maximized.
The grazing management of the experiment could also be changed by manipulating
livestock numbers and classes. During spring, the experimental stocking rate was 11.5
SU/ha. This equates to 9.6 ewes/ha plus their twin lambs. It could have been 4.4 Rising 1
(R1) beef heifers/ha, or a combination of the two. With 4.8 ewes/ha with twin lambs and
2.2 R1 beef heifers/ha, the sheep are likely to consume the leafy lucerne and grass first.
The cattle being less discriminate may eat lucerne stem and developing seed heads in the
grasses. Thus, a combination of both cattle and sheep could be utilised to maintain
pasture quality. This is an alternative technique to the leader and follower system.
Best management practices for farmers wanting to incorporate lucerne/grass mixes into
their farming system would be to learn how to manage a lucerne monoculture first. This
allows time to adjust to the grazing management required for lucerne stands before
adding grasses. The addition of grass to lucerne increases the intensity and difficulty of
grazing management, particularly due to the different growth patterns of each species.
However this study has shown that there are grass species like cocksfoot and brome that
can be sown with lucerne to complement its growth pattern. Further work may assess
the other species such as tall fescue but in all cases developing a grazing system to
maximize animal production while maintaining pasture quality will be the key to success.
83
6.2 Conclusions
•
Liveweight
production
was
greater
from
lucerne
monocultures
and
lucerne/cocksfoot than lucerne/brome.
•
Daily liveweight gains for stock were generally similar across all pastures.
However, stocking rates were different which resulted in differences in liveweight
production/ha.
•
There were no differences in dry matter yield across all pastures. This was a
results of no differences in water use, water use efficiency or thermal time growth
rates.
•
Pre- and post-graze botanical composition indicated that diet selection for
lucerne, and against sown grass and dead material, occurred especially in the
lucerne/brome pastures.
•
Grazing management of the lucerne/brome pasture was not effective in
maintaining pasture quality throughout the growing season.
•
There was no difference in ME and N content and yield between pastures.
Therefore, liveweight production differences were a result of stocking rate, not
the quantity or quality of herbage.
84
7
ACKNOWLEDGEMENTS
It’s hard to believe that I have finally made it this far, this dissertation always seemed like
a never ending project! There are so many people, whose support has made it possible to
complete this.
To my supervisor Professor Derrick Moot, your guidance and support has been amazing –
I am truly grateful to have had the opportunity to work with someone who has so much
knowledge. And yes, I have been converted to lucerne!
To Dr Anna Mills, for putting up with my constant door knocking and quick questions
which were anything but quick. This dissertation would not be anywhere near complete if
it wasn’t for your expertise in absolutely everything.
To Malcolm Smith, thank you for all the hard work that goes into data collection – it’s not
an easy task.
To my parents –Jan and Murray, your support and commitment throughout my four years
at Lincoln has been truly amazing. This degree is as much yours as it is mine!
To my flatmates – Mell, Cherie, Brooke and Kate – thank you for making this year so
enjoyable, and for all your help with data collection, it is much appreciated. To the field
service centre staff, your willingness to help has been much appreciated and made this
process so much easier.
And to my fellow honours students – we made it!
85
8
REFERENCES
Arnold, G. 1960. Selective grazing by sheep of two forage species at different stages of
growth. Crop and Pasture Science, 11, 1026-1033.
Avery, D., Avery, F., Ogle, G. I., Wills, B. J. and Moot, D. J. 2008. Adapting farm systems to
a drier future. Proceedings of the New Zealand Association, 70, 13-18.
Baars, J. and Cranston, A. 1978. Performance of ‘Grasslands Matua’prairie grass under
close mowing in the central North Island. Proceedings of New Zealand Grassland
Association, 39, 139-147.
Barker, D. J., Lancashire, J. A. and Meurk, C. 1985. 'Grasslands Wana' cocksfoot - an
improved grass suitable for hill country. Proceedings of New Zealand Grassland
Association, 46, 167-172.
Brown, H., Moot, D. and Pollock, K. 2003. Long term growth rates and water extraction
patterns of dryland chicory, lucerne and red clover. In: Legumes for dryland
pastures: proceedings of a New Zealand grassland association symposium, Lincoln
University, New Zealand. p 18-19.
Brown, H. E. and Moot, D. J. 2004. Quality and quantity of chicory, lucerne and red clover
production under irrigation. Proceedings of the New Zealand Grassland
Association, 66, 257-264.
Brown, H. E., Moot, D. J., Lucas, R. J. and Smith, M. 2006. Sub clover, cocksfoot and
lucerne combine to improve dryland stock production. Proceedings of the New
Zealand Grassland Association, 68, 109-115.
Brown, H. E., Moot, D. J. and Teixeira, E. I. 2005. The components of lucerne (Medicago
sativa) leaf area index respond to temperature and photoperiod in a temperate
environment. European journal of agronomy, 23, 348-358.
Cullen, N. 1965. A comparison of the yield and composition of various mixtures of lucerne
and grass sown in alternate rows with lucerne sown as a pure stand. New Zealand
Journal of Agricultural Research, 8, 613-624.
Douglas, G., Wang, Y., Waghorn, G., Barry, T., Purchas, R., Foote, A. and Wilson, G. 1995.
Liveweight gain and wool production of sheep grazing Lotus corniculatus and
lucerne (Medicago sativa). New Zealand Journal of Agricultural Research, 38, 95104.
Douglas, J. 1986. The production and utilization of lucerne in New Zealand. Grass and
forage science, 41, 81-128.
Douglas, J. and Kinder, J. 1973. Production and composition of various lucerne and grass
mixtures in a semi-arid environment. New Zealand Journal of Experimental
Agriculture, 1, 23-27.
Edwards, G. R., Lucas, R. J. and Johnson, M. R. 1993. Grazing preference for pasture
species by sheep is affected by endophyte and nitrogen fertility. Proceedings of
the New Zealand Grassland Association, 55, 137-141.
Fasi, V., Mills, A., Moot, D. J., Scott, W. R. and Pollock, K. 2008. Establishment, annual
yield and nitrogen response of eight perennial grasses in a high country
environment. Proceedings of the New Zealand Grassland Association, 70, 123-130.
Fraser, T. 1982. Evaluation of ‘Grasslands Matua’prairie grass and ‘Grasslands
Maru’phalaris with and without lucerne in Canterbury. New Zealand Journal of
Experimental Agriculture, 10, 235-237.
86
Fraser, T., Moss, R., Daly, M. and Knight, T. 1999. The effect of pasture species on lamb
performance in dryland systems. Proceedings of the New Zealand Grassland
Association, 61, 23-30.
Fraser, T. J. and Vartha, E. W. 1979. Experience with lucerne-grass systems for sheep
production. Proceedings of the New Zealand Grassland Association, 41, 50-55.
Grant, S., Suckling, D., Smith, H., Torvell, L., Forbes, T. and Hodgson, J. 1985. Comparative
studies of diet selection by sheep and cattle: the hill grasslands. The Journal of
Ecology, 987-1004.
Haynes, R. and Williams, P. 1993. Nutrient cycling and soil fertility in the grazed pasture
ecosystem. Advances in agronomy, 49, 119-199.
Joyce, J. and Brunswick, L. 1975. Sodium supplementation of sheep and cattle fed
lucerne. New Zealand journal of experimental agriculture, 3, 299-304.
Keogh, R. 1986. Fungal distribution and livestock defoliation patterns in pasture
ecosystems, and the development and control of dietary-dependent disorders.
Proceedings of the New Zealand Grassland Association, 47, 93-98.
Kirsopp, S. 2001. Management techniques to maximise legume production in dryland
farming. Masters of Applied Science (MSc) Thesis, Lincoln University, Canterbury,
New Zealand.
Langer, R. and Keoghan, J. 1970. Growth of lucerne following defoliation. Proceedings of
the New Zealand Grassland Association, 32, 98-107.
Lucas, R. J., Smith, M., Jarvis, P., Mills, A. and Moot, D. J. 2010. Nitrogen fixation by
subterranean and white clovers in dryland cocksfoot pastures. Proceedings of the
New Zealand Grassland Association, 72, 141-146.
Marsh, K. and Brunswick L.F.C. 1977. Beef production from lucerne and lucerne/prairie
grass swards on the pumice soils of the Taupo region. Proceedings of N, 39, 79-85.
McKenzie, B., Gyamtsho, P. and Lucas, R. J. 1990. Productivity and water use of lucerne
and two lucerne-grass mixtures in Canterbury. Proceedings of the New Zealand
Grassland Association, 52, 35-39.
McKenzie, B. A., Kemp, P. D., Moot, D. J., Matthew, C. and Lucas, R. J. 1999.
Environmental effects on plant growth and development In: J. White and J.
Hodgson (eds). New Zealand Pasture and Crop Science. Victoria Oxford University
Press, 35.
McLenaghan, R. and Webb, T. 2012. Soil properties and fertility of Ashley Dene: the
challenge In. Ashley Dene Lincoln University Farm. The First 100 Years.
Christchurch: Christchurch Digital Print
Mills, A. and Moot, D. J. 2010. Annual dry matter, metabolisable energy and nitrogen
yields of six dryland pastures six and seven years after establishment. In:
Proceedings of the New Zealand Grassland Association. Vol. 72. p 177-184.
Mills, A., Moot, D. J. and McKenzie, B. A. 2006. Cocksfoot pasture production in relation
to environmental variables. Proceedings of the New Zealand Grassland
Association, 68, 89-94.
Mills, A., Smith, M., Lucas, R. J. and Moot, D. J. 2008a. Dryland pasture yields and
botanical composition over 5 years under sheep grazing in Canterbury.
Proceedings of the New Zealand Grassland Association, 70, 37-44.
Mills, A., Smith, M. C. and Moot, D. J. 2008b. Sheep liveweight production from lucerne,
cocksfoot or ryegrass based pastures. Proceedings of the 14th ASA Conference, 2125 September 2008, Adelaide, South Australia, 1-4.
87
Moot, D., Robertson, M. and Pollock, K. 2001. Validation of the APSIM-Lucerne model for
phenological development in a cool-temperate climate. Australian Agronomy
Conference, 10.
Moot, D. J. 2012. An overview of dryland legume research in New Zealand. Crop &
Pasture Science, 63, 726-733.
Moot, D. J., Brown, H. E., Pollock, K. and Mills, A. 2008. Yield and water use of temperate
pastures in summer dry environments. Proceedings of the New Zealand Grassland
Association, 70, 51-57.
Nicol, A. M. and Brookes, I. M. 2007. The metabolisable energy requirements of grazing
livestock. In: P. V. Rattray, I. M. Brookes and A. M. Nicol (eds). Pasture and
supplements for grazing animals. Christchurch: Printmax, 151-172.
O'Connor, K. F. 1967. Sociability of lucerne. In: R. H. M. Langer (ed). The Lucerne Crop.
Wellington A.H. & A.W. Reed, 47-61.
Peri, P. L., Moot, D. J., McNeil, D. L., Varella, A. C. and Lucas, R. J. 2002. Modelling net
photosynthetic rate of field-grown cocksfoot leaves under different nitrogen,
water and temperature regimes. Grass and Forage Science, 57, 61-71.
Sherrell, C. 1984. Sodium concentration in lucerne, phalaris, and a mixture of the 2
species. New Zealand Journal of Agricultural Research, 27, 157-160.
Sim, R., Moot, D., Brown, H. and Teixeira, E. 2012. Development, growth and water
extraction of seedling lucerne grown on two contrasting soil types. In: Capturing
Opportunities and Overcoming Obstacles in Australian Agronomy. Proceedings of
the 16th Australian Agronomy Conference
Tonmukayakul, N., Moot, D. and Mills, A. 2009. Spring water use efficiency of six dryland
pastures in Canterbury. Agronomy New Zealand, 39, 81-94.
Trafford, G. and Trafford, S. (eds). 2011. Farm Technical Manual. Christchurch: The
Caxton Press.
Vartha, E. 1975. Comparative annual and seasonal growth of three ryegrass varieties and
cocksfoot at Lincoln, Canterbury. New Zealand journal of experimental agriculture,
3, 319-323.
Vartha, E. W. 1973. Performance of lucerne-grass pastures on Wakanui silt loam. New
Zealand Journal of Experimental Agriculture, 1, 29-34.
Waghorn, G. C. and Barry, T. N. 1987. Pasture as a nutrient source. In: A. M. Nicol (ed).
Feeding livestock on pasture. Occasional publication No. 10, New Zealand Society
of Animal Production, 21-38.
White, J. G. H. 1982. Lucerne grazing management for the 80s. . In: R. B. Wynn-Williams
(ed). Lucerne for the 80s. Palmerston North: Agronomy Society of New Zealand,
111-114.
Wigley, K., Moot, D. J., Khumalo, Q. and Mills, A. 2012. Establishment of lucerne (
Medicago sativa) sown on five dates with four inoculation treatments.
Proceedings of the New Zealand Grassland Association, 74, 91-96.
Wynn-Williams, R. B. 1982. Lucerne establishment - conventional. In: R. B. Wynn-Williams
(ed). Lucerne for the 80's: Agronomy Society of New Zealand, Special publication
No. 1, 11-19.
88
9
APPENDICES
Appendix 1 Daily metabolisable energy requirements (MJ ME) of ewes for maintainance
and various stages of preganancy and lactation. From Nicol and Brookes
(2007).
The metabolisable energy requirement for maintenance of adult ewes.
Class
40
Flat land
Rolling/easy hill
Hard hill
7.5
Liveweight (kg)
60
70
MJ ME/ewe/day
9.0
10.0
8.0
10.0
11.0
9.0
11.0
50
80
11.0
The metabolisable energy requirement of ewes for pregnancy (in addition to
maintenance requirement).
Lamb birth
weight (kg)
-6
3
4
5
6
1.5
2.0
2.5
3.0
Weeks before lambing
-4
-2
MJ ME/ewe/day
2
3
3
4
3.5
5
4.5
6
0
4.5
6
7
8.5
Total for
pregnancy
MJ ME
155
200
255
300
The metabolisable energy requirements of ewes and their lambs during lactation (in
addition to ewe maintenance requirement).
Lamb weaning
weight (kg)
20.0
25.0
30.0
35.0
Weeks after lambing
Total for
2
6
10
12
lactation
MJ ME
MJ ME/ewe plus lamb(s)/day
8.5
10.5
12.5
13.0
855
10.5
13.0
16.0
17.0
1075
12.0
16.0
20.0
21.0
1335
14.5
19.5
24.5
26.0
1625
89
Appendix 2 Soil map of paddocks C6E, C7W and C7E at Ashley Dene, Canterbury, New
Zealand.
90
Appendix 3 Detailed stock movements for grazing rotations on lucerne monocultures,
lucerne/cocksfoot and lucerne/brome mixes at Ashley Dene, Canterbury, New
Zealand from 1/07/12 to 30/06/13.
Summary of grazing periods for stock grazing lucerne plots in paddocks C6E, C7W and C7E, Ashley
Dene, Canterbury. Ewes and lambs are represented by ‘E and L’, Ram hoggets are ‘Ram Hgts’ and
ewe hoggets are ‘Ewe hgts’.
Rotation
1
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
3
3a
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
4
4
4
4
4
4
5
5
5
5
5
5
5
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Plot
1
1
16
5
14
9
12
1
16
5
14
9
12
1
16
5
14
9
12
12
1
1
5
9
12
14
16
1
5
9
12
14
16
9
12
1
5
5
14
16
9
12,1
5
14
16
Stock
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
WL
WL
WL
WL
WL
WL
WL
WL
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
13
57
57
57
57
57
57
57
57
57
57
57
57
-
Number of stock
Lambs
Other
26
106
106
106
106
106
106
106
106
106
106
106
106
106
106
106
106
106
106
75
61
150
105
200
105
200
150
157
78
210
210
210
210
51
51
69
69
85
85
85
190
190
165
165
165
Date on
5/09/2012
18/09
21/9
26/09
3/10
8/10
16/10
24/10
31/10
5/11
12/11
17/11
24/11
28/11
10/12
17/12
24/12
29/12/2012
5/01
8/01
11/01
22/01
22/01
18/01
18/01
25/01
25/01
5/03
20/03
10/03
14/03
19/03
23/03
15/05
24/05
30/05
7/06
12/06
17/06
22/06
3/06
6/06
20/06
24/06
27/06
Date off
18/08
21/09
26/09
3/10
8/10
16/10
24/10
31/10
5/11
12/11
17/11
24/11
28/11
10/12
17/12
24/12
29/12
5/01/2013
8/01
11/01
19/01
25/01
25/01
22/01
22/01
28/01
30/01
8/03
26/03
13/03
16/03
21/03
24/03
24/05
30/05
7/06
12/06
17/06
22/06
26/06
4/06
10/06
21/06
25/06
28/6
91
Summary of grazing periods for stock grazing lucerne/cocksfoot plots in paddocks C6E, C7W and
C7E, Ashley Dene, Canterbury.
Rotation
1
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
3
3a
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
4
4
4
4
4
4
5
5
5
5
5
5
5
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Plot
2
2
18
4
13
8
10
2
18
4
13
8
10
2
18
4
13
8
10
10
2
2
4
8
10
13
18
2
4
8
10
13
18
8
10
2
4
4
13
18
8
10
2
4
13
18
Stock
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
WL
WL
WL
WL
WL
WL
WL
WL
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
14
56
56
56
56
56
56
56
56
56
56
56
56
-
Number of stock
Lambs
Other
28
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
77
66
200
105
200
150
150
105
78
78
210
142
210
210
54
54
70
70
89
89
89
190
190
190
165
165
165
Date on
5/09/2012
18/09
21/9
26/09
3/10
8/10
16/10
24/10
31/10
5/11
12/11
17/11
24/11
28/11
10/12
17/12
24/12
29/12/2012
5/01
8/01
11/01
22/01
24/01
18/01
18/01
28/01
28/01
8/03
14/03
8/03
8/03
18/03
26/03
15/05
23/05
30/05
7/06
12/06
17/06
22/06
1/06
4/06
10/06
17/06
23/06
30/06
Date off
18/08
21/09
26/09
3/10
8/10
16/10
24/10
31/10
5/11
12/11
17/11
24/11
28/11
10/12
17/12
24/12
29/12
5/01/2013
8/01
11/01
19/01
25/01
28/01
22/01
23/01
2/02
4/02
14/03
20/03
10/03
10/03
19/03
28/03
23/05
30/05
7/06
12/06
17/06
22/06
26/06
3/06
5/06
13/06
20/06
24/06
3/07
92
Summary of grazing periods for stock grazing lucerne/brome plots in paddocks C6E, C7W and C7E,
Ashley Dene, Canterbury.
Rotation
1
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
3
3a
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
Cleanup 1
4
4
4
4
4
4
5
5
5
5
5
5
5
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Cleanup 2
Plot
3
3
17
6
15
7
11
3
17
6
15
7
11
3
17
6
15
7
11
11
3
3
6
7
11
15
17
3
6
7
11
15
17
7
11
3
6
6
15
17
7
11
3
6
15
17
Stock
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
E&L
WL
WL
WL
WL
WL
WL
WL
WL
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ram Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewe Hgts
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
Ewes
14
55
55
55
55
55
55
55
55
55
55
55
55
-
Number of stock
Lambs
Other
28
98
98
98
98
98
98
98
98
98
98
98
98
98
98
98
98
98
98
67
42
200
150
150
105
200
105
142
210
210
210
210
210
55
55
65
65
89
89
89
190
190
190
165
165
165
Date on
5/09/2012
18/09
21/9
26/09
3/10
8/10
16/10
24/10
31/10
5/11
12/11
17/11
24/11
28/11
10/12
17/12
24/12
29/12/2012
5/01
8/01
11/01
22/01
22/01
18/01
18/01
25/01
25/01
5/03
20/03
10/03
14/03
19/03
23/03
15/05
24/05
30/05
7/06
12/06
17/06
22/06
3/06
5/06
13/06
21/06
25/06
28/06
Date off
18/08
21/09
26/09
3/10
8/10
16/10
24/10
31/10
5/11
12/11
17/11
24/11
28/11
10/12
17/12
24/12
29/12
5/01/2013
8/01
11/01
19/01
25/01
25/01
22/01
22/01
28/01
30/01
8/03
26/03
13/03
16/03
21/03
24/03
24/05
30/05
7/06
12/06
17/06
22/06
26/06
4/06
6/06
16/06
23/06
27/06
30/06
93
Appendix 4 Regression equations and coefficients of determination for the regression of
cumulated thermal against cumulated dry matter yield time in spring 2012 of
lucerne monocultures, lucerne/cocksfoot and lucerne/brome mixes at Ashley
Dene, Canterbury, New Zealand.
Plot
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Equation
5.43x -1366
5.06x -1102
5.50x - 1408
5.60x -1470
5.73x - 1689
5.90x - 1828
5.22x - 2892
5.15x - 2150
5.45x - 2353
5.32x - 2079
5.44x - 2550
5.21x - 2056
5.52x - 2223
5.69x - 2099
5.71x - 2387
5.65x - 1814
5.62x - 1920
5.44x - 1901
R2
0.98
0.99
0.98
0.99
0.99
0.99
0.99
0.99
0.99
0.98
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
94
Appendix 5 Regression equations and coefficients of determination for the regression of
cumulated water use (WU) against cumulated dry matter yield in spring 2012
of lucerne monocultures, lucerne/cocksfoot and lucerne/brome mixes at
Ashley Dene, Canterbury, New Zealand.
Plot
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Equation
20.77x + 762
19.46x + 876
20.99x + 750
21.76x + 724
22.91x + 457
21.39x + 719
20.73x - 874
22.69x - 579
24.06x - 682
20.08x + 139
23.07x - 803
22.14x - 407
22.53x - 189
24.20x - 159
24.98x - 556
21.79x + 430
21.35x + 351
21.11x + 241
R2
0.991
0.996
0.992
0.995
0.999
0.999
0.996
0.997
0.995
0.999
0.995
0.999
0.999
0.999
0.998
0.999
0.997
0.999
95
Appendix 6 Water extraction pattern of lucerne and brome roots in the soil profile.
Where (●) is the upper limit and (○) is the lower limit for plant available
water in plots 1,3-6,8-18 in paddocks C6E, C7W and C7E at Ashley Dene,
Canterbury, New Zealand.
96
97
98
99
100
101
102
103